Autonomy: From Myth to Reality – or Hubris on a Tightrope?EU Law, Human Rights and International Law

Katja S Ziegler*

This chapter considers the relationship between EU law and international law from the perspective of human rights in regard to both EU fundamental rights and international human rights, in particular the European Convention on Human Rights. It introduces its theme by considering the characteristics, links and contrasts of EU fundamental rights in relation to international human rights (1). It then analyses the approach of the Court of Justice of the European Union to the interaction between EU law, including EU Fundamental Rights, and international law generally as moving from an open to a restrictive approach (2). More specifically, the relationship between EU law and international human rights, in particular the ECHR, will then be considered (3). It is shown and critically analysed how the CJEU has increasingly focused on the autonomy of the legal order, not just in relation to the Member States, but also in relation to international law up to the point of constructing a notion of autonomy akin to state sovereignty (4). The concluding section, not the least because of persisting human rights deficits of the EU and their impact on the legitimacy and credibility of the EU, suggests developing an alternative approach to general international law, informed by the rule of law and openness to international law to promote good governance (5).

1. EU fundamental rights and international human rights law: some characteristics, links and comparisons

The relationship between EU fundamental (or human)1 rights and international human rights is multifaceted:

First, EU fundamental rights are themselves international human rights. Regardless how far along the road of a sui generis nature and/or towards statehood one may consider the EU to have travelled, as long as the EU is not a state EU fundamental rights will be part of the body of international (human rights) law. EU fundamental rights also contribute to the development of international and constitutional human rights law.2

Secondly, the link between international and EU human rights is, as is well known, strongest at the point of genesis of EU human rights. In the light of an existential necessity, given the challenge put on by national constitutional courts because of the lack of fundamental rights protection in the European Communities, the CJEU developed the body of unwritten human rights law through the vehicle of general principles of EU law. In doing so, it borrowed extensively and substantively from international human rights (in particular the ECHR), as well as constitutional rights of its Member States.3

1* I am grateful to Richard Davy for his comments on this chapter. Used synonymously in this chapter, acknowledging that the adjective ‘fundamental’ tends to have more constitutional connotations.2 See Benkharbouche and Janah v Embassy of the Republic of Sudan and Libya [2015] EWCA Civ 33 in regard to state immunity and a right to a remedy. Katja S Ziegler, ‘Immunity v Human Rights?Incompatibility of the State Immunity Act with the HRA and the Right to a Remedy under International and European Law after Benkharbouche’ (forthcoming).3 See in detail Chapter XXX in this volume CROSSREFERENCE???

1

Thirdly, there is also one aspect which is specific to EU human rights, in particular in comparison with other international human rights regimes. Because the EU operates on the principle of conferred competence, the structural limitations of EU law also mean that although EU human rights restrict EU action, there is so far no comprehensive human rights competence of the EU. In other words, despite their significance and potential impact and the advanced degree of constitutionalisation of the EU also in the quasi-federal (or vertical) dimension in regard to Member States, EU fundamental rights are still less ambitious and less all-encompassing today than virtually all other protections of human rights at international level where targets are States. Human rights protection in the EU today also contrasts with the protection envisaged originally for the European Political Community (EPC), as de Búrca has shown,4 and the human rights protection by international treaties for the protection of human rights, such as the ECHR and the International Covenants on Civil and Political Rights (ICCPR) and on Economic Social and Cultural Rights (ICESCR). EU fundamental rights are, in contrast to other international systems of protection, in principle not intended to govern the relationship with the Member States (see in particular Article 51(2) of the EU Charter of Fundamental Rights). This peculiarity, together with the connected issue of a limited human rights competence of the EU,5 causes some inherent frictions and inconsistencies in EU law. It also creates some tensions between EU law and international human rights law, in particular in regard to the principle of non-discrimination and equality.6

Fourthly, related to the previous point: the relationship with the Member States is one factor that influences (and complicates) the approach of the EU to international law, making the relationship a triangular one.7

Fifthly, the status of EU fundamental rights in the constitutional hierarchy of the EU may cause a potential clash with international law: the CJEU, starting with Kadi I and put in the foreground of its reasoning in Kadi II, adopted an approach based on constitutional values reflected in EU fundamental rights8 which did not allow derogations. It thus gave EU fundamental rights a higher status than ordinary treaty rules, which can be derogated under certain circumstances.

Sixthly, the relationship between EU fundamental rights and international human rights is also ambivalent in regard to the level of protection EU fundamental rights provide or are able to provide. On the one hand the body of EU fundamental rights after the Charter is a modern system of human rights protection covering a wide variety of rights. On the other hand there are serious deficiencies in the protection of

4 Gráinne de Búrca, ‘The Road Not Taken: The European Union as a Global Human Rights Actor’ (2011) 105 AJIL 649.5 That this is not an inevitability is demonstrated by the illuminating study of totally different dynamics of economic integration in the context of ECOWAS by Karen J Alter, Laurence R Helfer and Jacqueline R McAllister, ‘A New International Human Rights Court for West Africa: The ECOWAS Community Court of Justice’ (2013) 107 AJIL 737.6 See Pälvi Johanna Neuvonen, Equal Citizenship and Its Limits in EU Law. We The Burden? (forthcoming Hart Publishing 2016).7 In more detail see Katja S Ziegler, ‘International Law and EU law: Between Asymmetric Constitutionalisation and Fragmentation’ in Alexander Orakhelashvili (ed), Research Handbook on the Theory and History of International Law (Edward Elgar Publishing 2011) 268. See also the theme of distrust towards the Member States that pervades recent Opinion 2/13 ECLI:EU:C:2014:2454.8 Joined Cases C-402&415/05 P Yassin Abdullah Kadi and Al Barakaat International Foundation v Council and Commission of the European Union [2008] ECR I-6351 (‘Kadi I’); C-285/12 Aboubacar Diakité v Commissaire général aux réfugiés et aix apatrides [2014] ECLI:EU:C:2014:39; C-308/06 Intertanko et al. v Secretary of State for Transport [2008] ECR I-4057.

2

human rights in the EU.9 These occur at different levels and for different reasons. The first group of reasons follows from the structural limitations of the limited competence of the EU or the deliberate exclusion of review by the Court in CFSP as a remnant of the pillar structure suggesting a lack of competence – for which ultimately the Member States are to blame. The second group of reasons relates to human rights deficiencies which result at the level of the rules – for which ultimately the EU and/or the CJEU are to blame. Such deficiencies can result from the interpretation of the Treaty rules, secondary legislation and its application. Examples are the restrictive interpretation of Article 263 TFEU in regard to the standing of individuals10 and environmental protection organisations11 and secondary legislation and its interpretation by the CJEU in the context of asylum and immigration law and justice and home affairs.12

As a result, the various dimensions of the relationship between EU fundamental rights and international law and international human rights in particular may give rise to two types of potential conflict (beyond a general conflict between EU law and international law) which may intersect:

1. The ‘constitutional’ dimension: there may be a clash of EU fundamental rights with general international law on the basis of more advanced human rights protection in the EU; and

2. The ‘compliance’ dimension: there may be a clash of EU law with international human rights law on the basis of limitations and gaps of human rights protection in the EU, which may mean the EU falls below the international minimum standard of protection.

Both dimensions are relevant for the EU, and both espouse legitimate concerns. It will be argued13 that an open approach to international law serves both dimensions and accommodates EU constitutionalism flexibly both on the evolutionary trajectory of the EU itself and in relation to the Member States. The recent trend of the case law of the CJEU is discussed critically against this backdrop.

2. EU fundamental rights and the interaction of EU law with international law: from openness to restriction

There are two dimensions to the question of interaction between EU law, including EU fundamental rights and international law: the interaction of EU law with international law generally and the interaction with international human rights more specifically. However, in both dimensions there is an increasingly discernible trend of the EU legal order closing up with regard to the international one. 9 Cf Andrew Williams, EU Human Rights Policies. A Study in Irony (Oxford University Press 2004).10 Case-583/11P Inuit Tapiriit Kanatami and Others v European Parliament and Council of the European Union, ECLI:EU:C:2013:625 paras 68 ff, 93 f.11 C-321/95 Stichting Greenpeace Council (Greenpeace International) and Others v Commission [1998] ECR I-1651; Joined cases C-404/12 P and C-405/12 P Council of the European Union and European Commission v Stichting Natuur en Milieu and Pesticide Action Network Europe [2015] ECR I-(nyr); Joined cases C-401/12 P to C-403/12 P Council of the European Union and Others v Vereniging Milieudefensie and Stichting Stop Luchtverontreiniging Utrecht, ECLI:EU:C:2015:5. .12 C-394/12 Shamso Abdullahi v Bundesasylamt ECLI:EU:C:2013:813; M.S.S. v Belgium and Greece App no 30696/09 (ECHR, 21 January 2011); Tarakhel v Switzerland App no 29217/12 (ECHR, 4 November 2014). Sionaidh Douglas-Scott, ‘The EU's Area of Freedom, Security and Justice: A Lack of Fundamental Rights, Mutual Trust and Democracy?’ (2008-9) 11 Cambridge Yearbook of European Legal Studies 53.13 See Section 5.

3

The treaty framework and the explicit self-understanding of the EU as expressed in political and legal texts, in particular in Article 3(5) TEU,14 foresees a general openness and embeddedness of the EU in the international legal order. Article 3(5) TEU constitutionally binds the EU to ‘strict observance and the development of international law, including respect for the principles of the United Nations Charter’. Moreover, and going beyond this, the constant jurisprudence of the CJEU refers to international law being an ‘integral part’ of EU law15 which has been interpreted to give rise to a presumption of direct applicability of international law within the EU legal order rather than merely constituting obligations at the international level.16 However, in spite of the general embeddedness of the EU in international law and an abstract approach to the relationship which suggests a general openness and friendliness towards international law, recent years have seen a number or cases in which the CJEU has taken a significantly more restrictive approach both to international law generally and to international human rights law more specifically in the EU legal order.

This section will first outline the general approach of the EU to international law. The approach to international human rights law more specifically will be examined in the next section (3).

2.1. Embeddedness of the EU in international law

It may be briefly recalled that the EU is embedded in a number of ways in the international legal order. Whatever sui generis nature one may accord to the EU by virtue of its substantive constitutionalisation, it remains formally an international organisation created by treaties, and its competence for independent secondary law-making is still derived from a treaty (Article 249 TFEU). Its international legal personality (Article 47 TEU) is still a derived one from the Member States, which

14 Article 3(5) TEU: ‘In its relations with the wider world, the Union shall uphold and promote its values and interests and contribute to the protection of its citizens. It shall contribute to peace, security, the sustainable development of the Earth, solidarity and mutual respect among peoples, free and fair trade, eradication of poverty and the protection of human rights, in particular the rights of the child, as well as to the strict observance and the development of international law, including respect for the principles of the United Nations Charter.’15 Case 181/73 R & V Haegeman v Belgium [1974] ECR 449, para 5; 104/81 Kupferberg v Hauptzollamt Mainz [1982] ECR 3641, para 13; C-162/96 A Racke & Co v Hauptzollamt Mainz [1998] ECR I-3633, para 55; Intertanko (n 8), para 38; C-366/10 Air Transport Association of America and Others v Secretary of State for Energy and Climate Change [2011] ECR I-13755, paras 73, 101 (‘ATAA’).16 Thus seems to be the prevailing opinion following the Court’s judgments in Haegeman and Kupferberg, above (n ); see also C-213/03 Pêcheurs de L'Étang de Berre v EDF [2004] ECR I-7357; C-377/98 Kingdom of the Netherlands v European Parliament and Council of the European Union (Directive on Biotechnological Inventions) [2001] ECR I-7079; C-213/03 Syndicat Professionnel Coordination des Pêcheurs de L'Étang de Berre v EDF [2004] ECR I-7357; C-344/04 Air Transport Association and European Low Fares Airline Association v Department for Transport (IATA) [2006] ECR I-403; Christian Tomuschat, ‘Artikel 281 EGV: Rechtspersönlichkeit der Gemeinschaft’ in Hans von der Groeben and Jürgen Schwarze (eds), Kommentar zum EU-/EG-Vertrag (6th edn, Nomos 2003), para 48; Christine Kaddous, ‘Effects of International Agreements in the EU Legal Order’ in Marise Cremona and Bruno de Witte (eds), EU Foreign Relations Law. Constitutional Fundamentals (Hart Publishing 2008) 291, 293 and n 11, 311; Anne Peters, ‘The Position of International Law within the European Community Legal Order’ [(1997) 40 German YIL 9, 21 ff, esp 34 f; recently: Mario Méndez, The Legal Effects of EU Agreements: Maximalist Treaty Enforcement and Judicial Avoidance Techniques (Oxford University Press 2013) 63, 268.

4

have transferred aspects of their sovereignty on the EU. The power to amend the treaties as well as to exit17 from the treaty system rests still with the Member States.

International personality means that the EU is a subject of international law which is bound by international law, but may also contribute to the creation of international law, both customary and treaty law (Article 216(1) TFEU, Article 37 TEU). The EU is party to innumerable treaties in its areas of competence,18 including mixed agreements where it has shared competence,19 and is thus both norm generator and norm recipient at international level. Additional obligations may arise because of the specific relationship of the EU and its Member States in regard to Member States’ international obligations into which the EU may functionally succeed.20

International law also applies by explicit or implicit cross-referencing of the EU legal order to concepts of international law (for example, nationality) and in the absence of leges speciales in the EU legal order as a gap-filler, i.e. the EU legal order is not a fully self-contained regime.21 Embeddedness of the EU legal order in international law is also reflected in the hierarchy of norms in EU law: international law, once directly applicable, ranks more highly than secondary legislation of the EU and thus may serve as a ground of review when reviewing the validity of EU secondary legislation.22

Two caveats have to be made in this regard. First, the degree of embeddedness in international law may change over time and with the evolution of EU law as already reflected in the self-understanding of the EU, as expressed by the CJEU.23 At the earlier stages of integration and constitutionalisation of the EU, international law seems to have been used to bolster the legitimacy of the EU and hierarchy of EU law in relation to the Member States (‘new legal order of international law’24). More 17 Articles 48 and 50 TEU limit the possibility of exit in so far as they bind states to a procedure.18 Piet Eeckhout, EU External Relations Law (2nd edn, Oxford University Press 2011) 71 ff.19 For more details about the specific problems relating to mixed agreements see, for example, Eleftheria Neframi, ‘Mixed Agreements as a Source of European Union Law’ in Enzo Cannizzaro, Paolo Palchetti and Ramses A. Wessel (eds), International Law as Law of the European Union (Martinus Nijhoff 2011) 325; Christophe Hillion and Panos Koutrakos (eds), Mixed Agreements Revisited: The EU and its Member States in the World (Hart Publishing 2010); Méndez (n 16).20 Joint Cases 21-24/72 International Fruit Co NV and Others v Produktschap Voor Groenten en Fruit [1972] ECR 1219, paras 110 ff.21 Bruno Simma and Dirk Pulkowski, ‘Of Planets and the Universe: Self-contained Regimes in International Law’ (2006) 17 European Journal of International Law 483, 505 f; Bruno de Witte, ‘European Union Law: How Autonomous is its Legal Order’ (2010) 65 Zeitschrift für Öffentliches Recht 141, 146; cf also Alain Pellet, ‘Les fondements juridiques internationaux du droit communautaire’ (1994) V-2 Collected Courses of the Academy of European Law 193, 203 f, who distinguishes different degrees of radicalism in regard to the relationship between international and EU law.22 Case C-162/96 A. Racke GmbH & Co v Hauptzollamt Mainz [1998] ECR I-3688, para 55; ATAA (n 15), para 50; Joint Cases C-335/11 and C-337/11 HK Danmark, acting on behalf of Jette Ring v Dansk almennyttigt Boligselskab, and HK Danmark, acting on behalf of Lone Skouboe Werge v Dansk Arbejdsgiverforening, acting on behalf of Pro Display A/S, in liquidation, ECLI:EU:C:2013:222, para 28; C-363/12 Z. v A. Government department and The Board of management of a community school, ECLI:EU:C:2014:159, paras 71, 84.23 See also Gráinne de Búrca, ‘International law before the Courts: The European Union and the United States compared’, NYU School of Law, Public Law Research Paper No 14-61. Available at SSRN: http://ssrn.com/abstract=2487361, forthcoming Virginia Journal of International Law (text relating to n 30 ff).24 Case 26/62 NV Algemene Transport- en Expeditie Onderneming van Gend & Loos v Netherlands Inland Revenue Administration [1963] ECR 2; 6/64 Flaminio Costa v E.N.E.L. [1964] ECR 587; see de Witte (n 21), 147 for an argument that the textual variation in Costa (dropping the reference to internaitonal law) did not reflect a change in approach of the CJEU regarding the embeddedness of the EU legal order in the international one.

5

recently the approach of a more constitutionalised EU seems to rely more on its own constitutional values, even against international law, stressing even the autonomy from international law.25 This marks a shift in paradigm and follows the logic of statehood. Opinion 2/13 of 18 December 2014 is the provisional high-water mark of such a potentially qualitative change of the nature of the EU.

Secondly, fundamental constitutional change aside, in the end what counts is not only an embeddedness of the EU in international law as an abstract principle, but how the relationship is tailored at the level of specific interaction. The CJEU is a gatekeeper in this regard. A principle of direct effect and rank of international law above-secondary legislation may fade into insignificance if, in practice, the ‘gatekeeping’ criteria of EU law determining when international law ‘enters’ EU law are very strict. This is, however, the increasing trend of CJEU jurisprudence.

2.2. The power to decide the form of interaction

The embeddedness of the EU in international law is normally assumed not to predetermine the form of interaction and status of international law within EU law. This is orthodoxy in regard to the interaction of international law within the legal order of states: international law is binding on states but it is up to states to decide whether and how it applies within them – or for agreements concluded by the EU, for the institutions of the EU negotiating and applying the agreement. As such the EU would be nothing special in this regard. However, the fact that the CJEU recently has stated this explicitly in Stichting Natuur en Milieu26 and Verenigign Milieudefensie27, stressing its power as a gatekeeper (even) in regard to treaties to which the EU is a party, seems significant as it is in line with a trend away from a presumption of direct effect of agreements to a stronger assertion of control by the CJEU.

‘However, the effects, within the EU legal order, of provisions of an agreement concluded by the European Union with non-member States may not be determined without taking account of the international origin of those provisions. In conformity with the principles of international law, EU institutions which have power to negotiate and conclude such an agreement are free to agree with the non-member States concerned what effects the provisions of the agreement are to have in the internal legal order of the contracting parties. If that question has not been expressly dealt with in the agreement, it is for the courts having jurisdiction in the matter and in particular the Court of Justice, within the framework of its jurisdiction under the FEU Treaty, to decide it, in the same manner as any other question of interpretation relating to the application of the agreement in question in the European Union on the basis in particular of the agreement’s spirit, general scheme or terms (see judgment in FIAMM and Others v Council and Commission, C-120/06 P and C-121/06 P, EU:C:2008:476, paragraph 108 and the case-law cited).’28

25 Kadi I (n 8), Diakité (n 8), Intertanko (n 8); C-465/07 M. and N. Elgafaji v Staatssecretaris van Justitie [2009] ECR I-921, para 28.26 Stichting Natuur en Milieu (n 11), para 45.27 Vereniging Milieudefensie (n 11), para 53.28 Stichting Natuur en Milieu (n 11), para 45.

6

While the issue cannot be analysed in greater detail here, some arguments which point in the direction of an erosion of the orthodoxy of the power to determine the interaction,29 derived from an analogy with national constitutional law, deserve mention here. These arguments highlight that the CJEU may be clinging to an increasingly out-dated conception of the relationship between international and domestic law.

Firstly, as a matter of international law, this dualist premise for the interaction has recently been questioned even in regard to states as original subjects of international law on several levels.30 The teleological argument, which ultimately can be derived from the pacta sunt servanda principle of international law,31 that direct application minimises breaches of international law, as well as a construction that focuses on individual state institutions being bound by international law, are not new. However, they have gained strength from comparative analyses of approaches of courts making normative choices about the interaction32 more along monist lines, as well as from an acknowledgment that direct effect ought to be established by interpretation of the international norm.33

Secondly, and directly following from the first point, it may be said that in an EU context, the rule that Member States determine the effect of international law (and therefore EU law) in their legal system has been eroded. It was in fact the CJEU (at international law level) which adopted a wide approach to direct effect of EU law (as a type of international law) in the Member States’ legal orders, applying a strong teleological interpretation of EU law on the basis of its effectiveness (effet utile) and uniformity of application. So there is a tradition of defining the interaction in the national legal orders of the Member States at international (EU) level. If that principle is applied consistently, an argument may be made that general international law would interact in the same way with the EU legal order. Although this would be a plausible argument it is not an inevitable logical consequence. There would be at least two possible counterarguments in an EU-specific context: that the Member States have actively taken the decision in the treaties to restrict the applicability of international law in the EU legal order, which would be relatively weak given the absence of any specific rule in the treaties; and the stronger argument of the specificity of EU law as a new legal order of international law. Whereas the specific interaction between international law and EU law as a matter of EU law is arguable, the more general point still stands that the way the CJEU has in fact shaped the interaction of EU law with the Member States somewhat weakened the traditional proposition that international law is totally neutral in regard to its effect within ‘domestic’ legal orders (i.e. states and by extension the EU itself).

Thirdly, also as a matter of international law, it may be asked whether the fact that the EU is an international organisation is relevant or whether any state paradigm as to the approach to international law can simply be transposed. In other words, is the

29 Gerrit Betlem and André Nollkaemper, ‘Giving Effect to Public International Law’ (2003) 14 European Journal of International Law 569, 573.30 André Nollkaemper, ‘The Duality of Direct Effect in International Law’ (2014) 25 European Journal of International Law 105; Betlem and Nollkaemper (n 29).31 Codified in Art 26 of the Vienna Convention on the Law of Treaties (VCLT). Its wider formulation in regard to customary law would be a rule that whatever is customary law is binding.32 See for overview Eileen Denza, ‘The Relationship between International and National Law’ in Malcolm D. Evans (ed), International Law (Oxford University Press 2010) 411, 418 ff; and for EU Member States specifically Méndez (n 19) 21 ff (but concluding, eg p 58 f, that the interaction is still largely determined at domestic level). 33 Nollkaemper (n 30) at notes 6 and around 12, 93, 98.

7

fact that the EU is a creature of international law conclusive for an argument of maximum openness if the treaties are silent on the issue?

Fourthly, not as a matter of international law but of EU law, the relationship of the EU with its Member States is particularly relevant and may be even more determinative than international law itself. A presumption that the treaties intend to avoid conflicts between EU and Member States’ obligations, a presumption of direct effect that can be derived from the principle of sincere cooperation in Article 4(3) TEU, further supports such a responsibility in order to avoid double-bind situations for the Member States.

Fifthly, from a more general, including a comparative, perspective it may be said that the general trend, at least in the area of human rights, is one of legal orders opening up to the consideration of not only international norms but also foreign law – i.e. other jurisdictions’ approaches to human rights issues.34 A closing up of the EU legal order thus runs counter to a trend visible at state level.

2.3. The trend towards a more closed relationship of the EU with international law in the mechanisms of interaction – as defined by the CJEU

In spite of the embeddedness and rootedness in international law, the relationship of the EU with international law has become increasingly less open. Such a trend is visible both in regard to treaties and customary international law and in regard to the way international law is given effect in the EU legal order, i.e. through direct effect, consistent interpretation or a ‘substantive borrowing’ of the EU legal order from the international one.35 If directly effective within EU law, international law applies automatically without an act of transformation in the EU legal order. This is the strongest possible effect international law may have in EU law (and as EU law also in the legal order of the Member States, regardless of their own approach to international law). Absent direct effect, international law will still be relevant in interpreting EU law. Methodologically, the boundaries between direct effect and interpretation are sometimes blurred in the jurisprudence of the courts, in particular where the results of direct effect and interpretation are the same because an interpretation in conformity with international law has prevailed. Independently of direct effect or consistent interpretation, a ‘substantive borrowing’ may occur where no formal relationship exists, for example in order to fill gaps in EU law. International law may serve as persuasive authority and feed content into general principles of (EU) law, in which case it also serves the interpretation of EU law, for example, as was the case for EU human rights norms developed by the Court.36

Whereas until recently it could be said that there was a presumption of direct effect of both international treaties and customary law, the CJEU has been tightening the criteria significantly. Controlling the gateway criteria has become crucial.

First doubts whether the CJEU still adhered to the presumption for direct effect of treaties were raised by three decisions of 2008, Fiamm and Fedon, Intertanko and Kadi I.37 Whereas at the time it could still be queried whether they

34 The CJEU is, however, so far not part of such a trend, not even in regard to foreign approaches to the interpretaton of international treaties, de Búrca (n 23) (text nr note 62).35 See in more detail Ziegler (n 7). 36 Williams (n 9), 145-157; Philip Alston (ed) The EU and Human Rights (Oxford University Press 1999); Bruno de Witte, ‘International Law as a Tool for the European Union’ (2009) 5 European Constitutional Law Review 265; Anne Peters, ‘Supremacy Lost: International Law Meets Domestic Constitutional Law’ (2010) 3 Vienna Online Journal on International Constitutional Law 170.

8

already marked a trend of closure, in particular since Kadi II38 raised a glimmer of hope to the contrary, the CJEU has since continued in this direction.39 The three decisions of 2008 have been discussed extensively elsewhere.40 It may be recalled that traditionally treaties were presumed to be directly effective where

1. the EU was bound by the treaty rule; 2. the treaty rule was sufficiently clear, precise and unconditional to be capable

of direct application; and 3. direct effect was not precluded by the ‘nature and structure’ or ‘broad logic’ of

a treaty.41 All three gateway criteria of direct effect have been interpreted restrictively since

2008 in the post-Lisbon era. A higher threshold was applied to the core criteria of direct effect by requiring, in addition to the provision being clear, precise and unconditional, that a provision must confer a truly subjective/individual right (in relation to UNCLOS42 and the Kyoto Protocol43). Furthermore, in Glatzel and Z the Court took a very narrow approach to the basic criteria of direct effect (‘unconditional and sufficiently precise’), denying direct effect to the UN Convention on Disabilities,44 and as a result, denying it any significance in the case.45 This makes it much more difficult for individuals to enforce rights derived from international law which are frequently not cast in the language of individual rights. It also contrasts with the wider notion of individual rights and direct effect of EU law internally.46

More widespread use has been made of the exclusion from direct effect by virtue of ‘nature and structure’ of the treaty: traditionally the latter exception only was held to be present in one case, i.e. in regard to the GATT/WTO Agreement. This was considered to be exceptional and justified47 by the flexible nature of the obligations.48 37 Joined cases C-120/06 P and C-121/06 P FIAMM and Fedon v Council and Commission [2008] ECR I-6513; Intertanko (n 8), para 38; Kadi I (n 8).38 Joined Cases C-584/10 P, C-593/10 P and C-595/10 P European Commission et al v Yassin Abdullah Kadi,ECLI:EU:C:2013:518 (‘Kadi II’).39 For example, C-356/12 Wolfgang Glatzel v Freistaat Bayern, ECLI:EU:C:2014:350; Z. (n 22); C-481/13 Mohammad Ferooz Qurbani, ECLI:EU:C:2014:2101, paras 22 ff; Opinion 2/13 (n 7).40 Ziegler (n 7); Katja S Ziegler, ‘The Relationship between EU Law and International Law’ in Dennis Patterson and Anna Södersten (eds), Blackwell Companion to EU Law and International Law (Wiley-Blackwell 2015 forthcoming) <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2373296> accessed 22 July 2015. 41 Intertanko (n 8), para 45; Joined cases C-120/06 P and C-121/06 P FIAMM and Fedon v Council and Commission [2008] ECR I-6513 and ibid Opinion of Advocate General Maduro, para 27 ff; International Fruit Co NV v Produktschap Voor Groenten en Fruit (n 202), para 25 ff; Case C-192/89 S.Z. Sevince v Staatssecretaris van Justitie [1990] ECR I-3497; ATAA (n 15), para 53 ff.42 Intertanko (n 8), paras 59, 61. 43 ATAA (n 15), para 75 ff.44 Glatzel (n 39), para 69.45 Ibid 70 f: the weaker effect of harmonious interpretation could not overcome the clear wording of the EU Directive as it would have resulted in a contra legem interpretation; Z. (n 22), paras 87 f, 90. Cf HK Danmark (n 22), para 55.46 Cf in more detail Ziegler (n 40), 301 ff. In favour of the approach of the Court (albeit critical of the narrow interpretation of the subjective rights in the Intertanko case see Eeckhout (n 18), 382 ff.47 See however, the convincing criticism by Nikolaos Lavranos, ‘The Chiquita and Van Parys Judgments: An Exception to the Rule of Law’ (2005) 32 Legal Issues of European Integration 449–460; Pieter J. Kuijper and Marco Bronckers, ‘WTO Law in the European Court of Justice’ (2005) 42 Common Market Law Review 1313; Steve Peers, ‘Fundamental Right or Political Whim? WTO Law and the European Court of Justice’ in Gráinne de Búrca and Joanne Scott (eds), The EU and WTO (Hart Publishing 2001) 111.48 See n 41 above and in favour of this approach Eeckhout (n 18), 375 ff (more strongly in favour in the first edition of his book Piet Eeckhout, External Relations of the European Union (Oxford University Press 2004) 302 ff); Marco Dani, ‘Remedying European Legal Pluralism: The FIAMM and Fedon

9

Since 2008 the CJEU added to the list of structurally excluded treaties UNCLOS and the Kyoto Protocol49 and it extended the rationale to binding final decisions of the WTO Dispute Settlement Body (to which the reasons for excluding GATT/WTO Agreement from direct effect do not apply).50

While perhaps less significant if considered in isolation, it is significant in the overall context that the CJEU also took a restrictive approach to the question what obligations are binding on the EU in the first place – as it considers it as prerequisite of direct effect that the EU is bound by a treaty rule, which in principle is the case where the EU is party to a treaty. However, the EU is not infrequently in a position, in effect, to fulfil or ‘breach’ international obligations of its Member States without being a party to a treaty and formally bound. This has been flagged up in case law, for example, regarding the Chicago Convention on Civil Aviation,51 the UN Charter in regard to the EU’s implementation of sanctions imposed by the UN Security Council52 and also the Geneva Convention relating to the Status of Refugees.53 The approach of the CJEU therefore appears overly formalistic, in particular in its approach to the UN Charter.54 A suitable alternative approach would be possible in (slightly) adapting the CJEU’s principle of functional succession of the EU into Member States’ obligations – as a matter of EU law – by considering whether the EU has functionally succeeded into an individual obligation55 rather than requiring succession into the obligations of an entire agreement.56 The basis in EU law would lie in the fact that Member States, when creating the EU, cannot have intended to breach Article 103 of the UN Charter, which would be the result if EU law authorised the EU to act contrary to the UN Charter. More generally, the principle of sincere cooperation in Article 4(3) TEU could provide a similar basis for a succession of the EU into other pre-existing Member States’ obligations.

Direct effect of customary international law (CIL) which, therefore, in principle is also a standard for the validity review of EU secondary legislation,57 is still acknowledged in principle by the CJEU. But the principle of direct effect of CIL has been clouded in several ways. In addition to the well-known fact that the Court never annulled an EU act for breach of CIL,58 post-Lisbon it seems to have tightened

Litigation and the Judicial Protection of International Trade Bystanders’ (2010) 21 European Journal of International Law 303.49 Intertanko (n 8), paras 54, 58, 65; ATAA (n 15), para 75 ff.50 Eeckhout (n 18), 367 ff, 380 ff, expressing concern that some of the reasoning may be transferred to the ECHR/ECtHR once the EU accedes, at 381; Dani (n 48), 310.51 ATAA (n 15), para 69 ff.52 Kadi I (n 8), para 306 f.53 C-31/09 Nawras Bolbol v Bevándorlási és Állampolgársági Hivatal [2010] ECR I-5539; C-364/11 Mostafa Abed El Karem El-Kott and Others v Bevnándorlási és Állampolgársági Hivatal, ECLI:EU:C:2012:826, para 43; but contrast with Qurbani, paras 22 ff where the CJEU precluded consideration of the Refugee Convention via a narrow construction of the scope of its jurisdiction! Contrast with C-148, 149 & 150/13 A., B. and C. v Staatssecretaris van Veiligheid en Justitie, ECLI:EU:C:2014:2406 [2014] ECR I-(nyr), para 46 and Opinion of AG Sharpston, para 30.54 Although the approach has been somewhat mitigated in practice by the Court ‘supposing [the Charter] to be applicable.’ Kadi I (n 8), para 306 ff (emphasis added).55 Cf the CFI’s approach in T-315/01 Kadi v Council and Commission [2005] ECR II-3659, para 192 ff. Likewise Robert Uerpmann-Wittzack, ‘The Constitutional Role of International Law’ in Armin Von Bogdandy and Jürgen Bast (eds), Principles of European Constitutional Law (2nd edn, Hart, CH Beck and Nomos 2009) 131, 154 f; Ziegler (n 7), 288 f.56 International Fruit Co NV v Produktschap Voor Groenten en Fruit (n 22), paras 110 ff.57 Racke (n 15), para 55; T-115/94 Opel Austria GmbH v Council [1997] ECR II-39, para 77; ATAA (n 15), paras 74, 84, 107.58 Cf n 57 above.

10

the criteria of direct effect in ATAA. In ATAA the Court restricted both the criteria when individuals may rely on CIL and its level of review. CIL can only be invoked in regard to ‘first, those principles are capable of calling into question the competence of the European Union to adopt that act (…) and, second, [situations in which] the act in question is liable to affect rights which the individual derives from European Union law or to create obligations under European Union law in his regard’.59 Because ‘a principle of customary international law does not have the same degree of precision as a provision of an international agreement’, review by the court is limited to whether the EU legislative institutions have made ‘manifest errors of assessment concerning the conditions for applying those principles.’60

In this context the Opinion of the Advocate General in the case of Diakité61 is worth noting as contributing to doubts about the continued direct effect of CIL (in casu: IHL) in EU law, as well as to tightening further the criteria of consistent interpretation of EU law with international law. Although arriving at a convincing outcome by interpretation of EU law in the case of Diakité, which the Court followed, the Advocate General contributed unhelpful new elements to the debate of the relationship between EU and international law, raising doubts about the support of a continued presumption of direct effect of CIL from within the Court, and revealing some of the difficulties of consistent interpretation with international law.

The case in essence dealt with the definition of ‘serious harm’ which triggers subsidiary protection under the Qualification Directive of persons not qualifying for refugee status. Serious harm, in one variant at issue is defined by the Directive as ‘a serious and individual threat to a civilian’s life or a person by reason of indiscriminate violence in situations of international or internal armed conflict’ (Article 15 (c)). The CJEU adequately solved the dilemma through interpretation ‘taking into account the context’ of the provisions and thus indirectly referring to the rules of interpretation in Article 31(3) (c) of the Vienna Convention of the Law of Treaties (VCLT). It held that IHL (i.e. CIL to a large extent) and EU refugee law pursued different purposes, and therefore were distinctive areas of law, so that ‘internal conflict’ in the Qualification Directive did not have to have the same meaning as (customary) IHL. This seems, on the whole, a straightforward result, arrived at by interpretation, engaging with international law along the way, and, it may be noted, not by relying on autonomy62 of the EU legal order in any abstract or formal way.63 The outcome could perhaps have been backed up further by referring to other areas of international law (more relevant than IHL in this context, e.g. obligations under Article 3 ECHR64), but on the whole this approach by the Court and result seem appropriate. However, even

59 ATAA (n 15), para 107 (emphasis added). Eileen Denza, ‘International Aviation and the EU Carbon Trading Scheme: Comment on the Air Transport Association of America Case’ (2012) 37 European Law Review 314, 324 points to the problem with this approach as it potentially precludes relevant customary international law which is not directly effective as a standard of asssessment of the validity. 60 Racke (n 15), para 52; ATAA (n 15), para 110 (emphasis added).61 AG Mengozzi, Diakité (n 8).62 See in detail Section 4 below.63 In a similar vein see also Violeta Moreno-Lax, ‘Of Autonomy, Autarky, Purposiveness and Fragmentation. The Relationship between EU Asylum Law and International Humanitarian Law’ in David James Cantor and Jean-François Durieux (eds), Refuge from Inhumanity? War Refugies and International Humanitarian Law (Brill 2014) 295; Céline Bauloz, ‘The Definition of Armed Conflict in Asylum Law’ (2014) 12 Journal of International Criminal Justice 835, 844 f. 64 See however, Elgafaji v Staatssecretaris van Justitie (n 25), para 28, where the Court held that Article 15(c) of the Qualification Directive was to be interpreted ‘independently’ from Art 3 ECHR (which corresponeded to Art 15 (b) - which may explain why the CJEU in Diakité did not discuss Art 3 ECHR).

11

though not finding their way into the Court’s judgment, some points made by the Advocate General are cause for concern both in regard to direct effect of CIL in and for the scope of the duty of consistent interpretation of EU law:

He seems to make consistent interpretation conditional on ‘hermeneutic consistency’ between international law (IHL in this case) and EU law.65 Where this simply refers to interpretation and its limits, there is no problem with this concept. However, where it is elevated to a separate new principle or ‘gate keeping’ criterion, the explanation of the AG culminates in what may be considered a reversal of the rule-exception paradigm currently reflected in the Court’s approach to interpretation in line with international law. It is the ‘hermeneutic consistency’ (however defined) that needs to be justified in order to open the gates for consistent interpretation,66 and possibly also for the presumption of direct effect of CIL. If this is the case, consistent interpretation and the presumption of direct effect lose the character of being the general rule.67

Although the scenario in Diakité was not one of diverging applicable standards and concepts of international law and EU law, but simply one of interpretation of EU law, it reveals the dangers and confusions that can arise from a poorly drafted piece of EU legislation which is worded in a confusing and even misleading manner by using international law/IHL terminology in an unreflected way in the context of subsidiary protection.

The scope of consistent interpretation also seems in danger of becoming more restricted. Whereas Intertanko still reflected an approach that any rules of international law68 were to be used to interpret EU law, the Advocate General’s Opinion in Diakité seems to restrict such interpretation to rules binding the EU formally.69 International obligations of the Member States would therefore not be informing consistent interpretation.70 In other words, rules which may be considered to be ‘integral part’ of EU law would be narrowly constructed71 and potentially turn from an empirical (shorthand) description of an open relationship with international law in the status quo to an additional ‘gate keeping’ criterion, which would potentially eliminate the present distinction between direct effect and an interpretive effect of international law in the EU legal order.

As mentioned above, an independent requirement of ‘hermeneutic consistency’ could turn consistent interpretation with international law from rule to a specially justifiable exception, so that doubts are raised not just about the continued existence of a presumption of direct effect of international law in EU law but also about the presumption of consistent interpretation of EU law with international law.72

Finally, the method of ‘substantive borrowing’ from international law as a form of interaction between EU and the international legal order, including international human rights, has also been significantly reduced. This is in particular due to the fact that in the context of EU fundamental rights the ECHR has lost its

65 AG Mengozzi in Diakité (n 8), paras 27 ff.66 AG Mengozzi in ibid, para 71.67 In spite of the fact that AG Mengozzi acknowledges it to exist in general terms in paras 24 f of his Opinion. The concern about the reversal of a rule-exception relationship with international law have been heightened following Z (n 22), paras 82 ff68 Intertanko (n 8); Kadi I (n 8). 69 AG Mengozzi in Diakité (n 8), para 26.70 AG Mengozzi in ibid (n 8), para 26. 71 Cf Qurbani (n 53), para 22.72 Cf AG Mengozzi in Diakité (n 8), para 71. See for detailed analysis Moreno-Lax (n 63).

12

prominence to the Charter.73 But the wider attitude of the CJEU revealed in Opinion 2/13 on EU accession to the ECHR (discussed further in Section 4 below) points to further reasons, in particular a concern about the institutional position of the CJEU in the European human rights architecture, a profound distrust of the Court towards the EU Member States, and a worrying lack of understanding of the function of an external human rights supervision (and perhaps even of international law more generally).

Furthermore, given the fact that the ECtHR has itself come under attack (some even refer to legitimacy crisis of the Strasbourg system74), the fact that the CJEU has referred less to the ECHR post-Lisbon may not be due just to the greater prominence of the Charter. Substantive borrowing is a two-edged sword which can enhance legitimacy if the legal order from which the borrowing occurs is accepted and held in high esteem.

So far the discussion has focused on the approach of the EU to international law generally. The following section will consider the approach to international human rights as a more specific aspect of interaction. It may be thought that the approach to international human rights is characterised by greater openness than that to general international law, as human rights more directly reflect shared values and common principles to which a legal order subscribes than other areas of international law – and which therefore ‘migrate’ more easily between legal orders. However, the CJEU’s approach reflected in Opinion 2/13 tells a different story.

3. EU law and international human rights: a restrictive trend towards the ECHR

The focus of this section will be on the relationship of EU fundamental rights with the ECHR as the most prominent international human rights instrument in relation to the EU legal order. Although the Court occasionally refers to ICCPR or the Refugee Convention or the Convention on the Rights of the Child, such references are more sporadic and often not elaborated.75 In comparison to the references to the ECHR in the pre-Charter era there is lack of engagement with other international human rights instruments. In particular, the lack of external human rights supervision and gaps in the internal human rights protection of the EU and detaching EU human rights from international human rights developments has been viewed critically by international and human rights scholars.76 The lack of engagement with other international human rights instruments also explains the regrettable closure of the relationship with international law in the area of human rights in the Kadi I case of 2008 which sits at

73 See further Section 3 below, especially text around note 94.74 See for example ‘The European Court of Human Rights has neither Authority nor Legitimacy’ The Telegraph (London, 28 January 2012); cf Nick Huls, Maurice Adams and Jacco Bomhoff (eds), The Legitimacy of Highest Courts' Rulings (TMC Asser Press 2009); Janneke Gerards, ‘The Scope of ECHR Rights and Institutional Concerns’ in Eva Brems and Janneke Gerards (eds), Shaping Rights in the ECHR: The Role of the European Court of Human Rights in Determining the Scope of Human Rights (CUP 2013) 84, 86 ff. 75 See for example C-34/09 Gerardo Ruiz Zambrano v Office national de l’emploi (ONEm) [2011] ECR I-1177, where the Court did not include the ICCPR and CRC (which had been identified as relevant by AG Sharpston in her Opinion, para 9 f), in its judgment, but only mentioned the ECHR. The Refugee Convention is still cited relatively frequently as background of the EU’s asylum legislation (referred to by recitals of relevant directives). 76 Israel de Jesús Butler and Oliver de Schutter, ‘Binding the EU to International Human Rights Law’ (2008) Yearbook of European Law 277, 279 ff; Gráinne de Búrca, ‘After the EU Charter of Fundamental Rights: The Court of Justice as a Human Rights Adjudicator?’ (2013) 20 Maastricht Journal of European and Comparative Law 168, 173 f, 175; de Búrca (n 4), 680.

13

the fault-line of a heightened constitutional protection of human rights and the closure towards international law more generally, and international human rights in particular. The Court did not ‘interpret away’ (as it could have done) the conflict at the level of international law by recourse to international human rights and shared values of the international and the European legal orders.

This comparative lack of international human rights law in the jurisprudence of the CJEU is particularly astonishing, since all Member States are parties to a number of human rights instruments which would qualify as sources from which to develop general principles of law.77 This is particularly the case for the ICCPR and ICESCR78, the Convention on the Rights of the Child, the 1965 International Convention on the Elimination of All Forms of Racial Discrimination, the 1979 Convention on the Elimination of All Forms of Discrimination Against Women, the 1984 Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment. This may have its roots in a failure to appreciate the peculiar nature of, and perhaps lack of experience with and expertise in, international human rights protection.79 Butler and de Schutter give the example that the CJEU failed ‘to appreciate the nature and status of the ICCPR as well as the role of the HRC’,80 when, for example, revealing in Grant that it considered that the UN Human Rights Committee (HRC) ‘is not a judicial institution’, contrary to a generally accepted judicial (or at least quasi-judicial) function of the HRC.81

Whereas the close intertwining of the EU legal order with the ECHR through a process of substantive borrowing of general principles of law showed great openness to one corpus of international human rights, today even this relationship must be described as increasingly ambivalent, partly as a result of the EU’s own internal constitutionalism. The relationship has evolved over time and several phases may be distinguished. Here, the three-phase distinction between the pre- and the post-Charter era and a potential post-ECHR accession scenario shall suffice.

3.1. The relationship up to the entry into force of the EU Charter in 2009

The relationship between the EU and international human rights initially started out to be considered much more along the lines of international law being an integral part of the EU legal order with no clearly demarcated boundaries, akin to a monist construction of the relationship. In regard to human rights, and international human rights, there was also reluctance in the early years of the CJEU to subordinate EU law formally to such limits internally. A similar reluctance of the later Member States was manifested to some extent in the failing of the EPC, which would have achieved exactly this in a more comprehensive framework of European integration.82

From the 1970s human rights became a necessity for the EU legal order, given the challenge from national constitutional courts. Thus, the relationship between EU law and international human rights (in particular the ECHR) became a substantively open relationship. During that phase the EU ‘borrowed’ substance extensively from the ECHR, reaching a sort of equilibrium with the Member States, and making the ECHR the most important material source of virtually all human rights law in the EU, 77 See for ICCPR and ICESCR Opinion of AG Sharpston, C-73/08 Bressol and Others v Gouvernement de la Communauté francaise [2010] ECR I-2735 (GC), para 136. Butler and de Schutter (n 76), 284 f.78 See, for example, Bressol (n 77), paras 85 ff.79 de Búrca (n 76), 171. 80 Butler and de Schutter (n 76), 285.81 Ibid, 279 ff, referring to C-249/96 L. J. Grant v South West Trains Ltd [1998] ECR I-636, para 47.82 de Búrca (n 4), 653 ff.

14

developed via general principles of EU law, until the Charter entered into force in 2009. The ECHR has long been recognised by the CJEU as a treaty of ‘special significance’ in spite of the EU not being a party to it. The ECHR and the case law of the ECtHR must be ‘taken into consideration in interpreting that scope of that right in the Community legal order’.83

However, the relationship between the EU (or more precisely, the CJEU) and the ECHR as an international treaty, providing for a considerably constitutionalised part of international law,84 has not always been straightforward.

Firstly, the Court has explicitly stated that the ECHR was not formally binding and therefore also not directly applicable within the EU legal order,85 although at times the formal status appears to have been blurred, giving the impression that it was as good as binding and directly applicable. In the light of this de facto binding and directly applicable status of the ECHR via general principles of EU law and its significant contribution to EU human rights, it may be said that it was perhaps practically less pressing to make it also formally binding on EU institutions. This was the result of Opinion 2/94 which put an end to any attempts at accession of the EU to the ECHR until the Treaty of Lisbon created a specific provision requiring the EU to accede to the ECHR.86 It can be said that the equilibrium reached with the Member States formed the basis of the CJEU rejecting a formalisation of the open relationship with the ECHR by accession in Opinion 2/94.87 Nevertheless, the lack of formal status had disadvantages because of at least the potential for creating double-bind situations in terms of Member States’ obligations and, in practice more significantly, because it left considerable flexibility to the Court to define and apply EU human rights, and to do so without an external supervision.

Secondly, beyond formal status and the substantive human rights protection, the relationship seems to be characterised by considerations of competing jurisdictions and their relative hierarchies. These were brought fully in the open in the EU accession process to the ECHR,88 but are clearly visible already in Opinion 2/94. The rejection of a competence to accede thus appears to have been driven by concerns

83 Elgafaji v Staatssecretaris van Justitie (n 25), para 28.84 Although the legal order of international law created by the ECHR is a much more incomplete one than the EU in terms of subject matter, it is in comparison to other areas of international law highly constitutionalised, in particular because of the compulsory jurisdiction of the ECtHR, the individual complaints mechanism and the linked enforcement mechanism (supervision of judgments) in the Council of Ministers of the Council of Europe.85 C-571/10 Servet Kamberaj v Istituto per l’Edilizia sociale della Provincia autonoma di Bolzano (IPES), Giunta della Provincia autonoma di Bolzano, Provincia autonoma di Bolzano, ECLI:EU:C:2012:233, 59 ff, 63. The case concerned a situation where direct applicability would have led to an application of the ECHR to a Member State act, not an act of the EU. However, it is surprising, in particular, given the principle of non-discrimination at issue, that the CJEU did not discuss obligations resulting from general principles of EU law further (see para 60 of the judgment). AG Bot in his Opinion had held the ECHR to be directly effective (para 38), but the court not to have jurisdiction over the consequences of incompatibility of a Member State’s act with the ECHR (para 39). Opinion 2/13 (n 7), para 179 f. N.B. that the CJEU refers to the ECHR as not having ‘been formally incorporated into the legal order of the EU’ (para 180, emphasis added) which may be merely a glitch of language, meant to mean ‘binding’; the notion of ‘incorporation’ of treaties known to dualist systems does not fit the traditional monist approach to treaties of the EU. 86 Opinion 2/94 [1996] ECR I-1759.87 ibid. 88 For example in the attempt by the Court to cast the Bosphorus presumption into the accession agreement which seemed extraordinarily misguided in that it reflects a misunderstanding of the function of international human rights law and would have defeated the purpose of accession. See below text around n 116.

15

about its own future status, even though this might put potential strain on the equilibrium reached with the Member States.

This reluctance evidenced in Opinion 2/94 is all the more remarkable since the Court espoused a more expansive approach to competence in other contexts, and a justification of a competence to accede, as it seems, could have been possible with comparatively simple reasoning (but strong reasons): either on the basis of succession into Member States’ obligations in areas of EU competence;89 or on the basis of the EU accepting and shaping limits on its existing competences on the basis of general principles: the EU could within its existing competences subject itself to human rights (i.e. self-limiting its competence); and in doing so, it could also have acceded to the ECHR. The competence to do so would not be a human rights competence (which the EU still does not have after the Treaty of Lisbon), but it would have followed from the specific competences conferred on the EU. Human rights in their limiting function, not unlike the principle of proportionality, have nothing to do with the question of a (legislative) competence in the area of human rights (which would primarily relate to Member States’ acts).90

Of course practically things are never simple and are full of detail, as the lengthy EU accession process ultimately ending in failure has shown; that may even vindicate the position taken in Opinion 2/94 somewhat, in so far as it referred to the significance and intricacy at the level of detail EU accession to the ECHR would entail. However, also in this regard Opinion 2/94 was ultimately not convincing in that it based its reasoning on the ‘constitutional significance’ of accession ‘with fundamental institutional implications for the Community and for the Member States’.91 Even conceding that the accession would have made the ECHR formally directly applicable qua EU law also in the Member States, that effect would have been limited to the application of the ECHR to or within the scope of EU law. If anything, this would have been in the interest of the Member States fearing competence creep; hence arguments of the Court about protecting the Member States from the impact of accession seem to be partly misguided. As the substantive human rights protection afforded by the ECHR was beyond doubt,92 this leaves as the dominant reason for the position of the Court a concern for its own position in the judicial landscape of a wider Europe.

Opinion 2/94 was therefore legally not convincing but appears to have been driven by primarily institutional considerations relating to the jurisdiction of and institutional competition for the Court.93 It took the ‘masters of the treaties’ to (attempt to) correct that and try to oblige the EU to accede.

3.2. The relationship after the entry into force of the EU Charter in 2009

89 Because of the fragmented competence of the EU which would lead to a matching fragmented application of the ECHR (only where the EU has competence), this would presume that succession of the EU is possible into aspects of Member States’ obligations, not replacing them totally. This would require the CJEU to depart from its currently narrow reading of the succession into international treaty obligations of the Member States which requires a ‘full transfer of powers’, see Intertanko (n 8), para 49. See in greater detail critically Ziegler (n 7), 288 f.90 Similar see Eeckhout (n 18), 99 f.91 Opinion 2/94 (n 86), para 35.92 ibid, para 34: ‘Respect for human rights is … a condition of the lawfulness of Community acts.’93 ibid, para 34 seems to suggest this, referring to the ‘distinct international institutional system’ to which the EU would enter.

16

The entry into force of the Treaty of Lisbon, which made the EU Charter of Fundamental Rights binding, marks not just a further constitutionalisation of EU law and EU human rights law but also one of further closure towards international law and international human rights law more specifically. As has been shown, the CJEU’s references to the ECHR and to other international human rights instruments from which it substantively borrowed in the past have been dramatically reduced post-Lisbon,94 giving rise to concerns of becoming ‘conspicuously detached from other relevant sources of human rights law and jurisprudence.’95 The CJEU now tends to refer to the Charter rather than the ECHR.96 This is unsurprising: to an extent the legal force, convenience and visibility of the EU’s own bill of rights makes resort to external substantive sources largely superfluous; the alternative – to continue to apply general principles of EU law – would result in the additional onus to justify why ECHR rights are general principles. The Charter is the more immediately accessible, written source of EU human rights law. It is also a modernised bill of rights drawn from the constitutional traditions of the EU Member States which in some respects responds to changed needs (even though the ECHR itself has been developed dynamically). The focus on the domestic codification is in line with how several Member States approach international human rights law. Together with the fact that general principles have been under fire as a device to expand EU competence,97 it is to be expected that the Court should use the Charter as a formal source of human rights. The constitutional/statist paradigm on which the EU embarks is nevertheless noteworthy. It may not necessarily be a bad thing in regard to developing a robust ‘domestic’ human rights culture. The experience of the UK may show how the lack of an internalised human rights culture my cause fundamental problems for the protection of human rights, for example if it becomes acceptable to use human rights as scapegoats for unwanted decision making.98

The CJEU has also reiterated that the ECHR was not formally binding on the EU. However, what cannot be explained or justified is why, in the absence of other material in particular, the ECHR does no longer feature more prominently in the interpretation of the Charter rights99 which must be chiselled out for this relatively recent human rights instrument; moreover, Article 52(3) of the Charter stipulates that Charter rights corresponding to ECHR rights ‘shall be the same as those laid down by the said Convention.’ Even bearing in mind that until formal accession the ECHR is neither formally binding nor internally directly applicable, this is not what one would have expected, given the tradition of using the ECHR as a material source of EU law. Such a change in approach must therefore be considered as part of a regrettable trend of closure to the ECHR, international human rights and international law. While the reasons for this can only be a matter for speculation, it could be asked 94 de Búrca (n 76), 171, f, 173-176.95 ibid, 183.96 ibid, 175 f and n 25.97 In particular since the debate about a potentially too expansive approach to general principles triggered by C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-10013 which was followed by a more cautious appraoch, e.g. C-13/05 Sonia Chacón Navas v Eurest Colectividades SA [2006] ECR I-6488; C-354/13 Fag og Arbejde (FOA), acting on behalf of Karsten Kaltoft v Kommunernes Landsforening (KL), acting on behalf of the Municipality of Billund, ECLI:EU:C:2014:2463, paras 32 ff, 40.98 See Katja S Ziegler, Elizabeth Wicks and Loveday Hodson (eds), The UK and European Human Rights: A Strained Relationship? (Hart Publishing 2015).99 Cf Elgafaji v Staatssecretaris van Justitie (n 25), para 28 which stipulated that case law of the ECHR and its interpretation by the ECtHR had to be ‘taken into consideration in interpreting that scope of that right in the Community legal order’.

17

whether these are the early signs of a more restrictive approach to the ECHR in a post-accession scenario, modelled on the traditional approach of some Member States which do not apply the ECHR directly as a standard of human rights review.100 Germany would be a case in point where the Federal Constitutional Court (FCC) traditionally held that only the human rights of the Grundgesetz are direct standard of review,101 whereas the ECHR informs their interpretation, thus avoiding breaches of the Convention. However, it has to be noted that the FCC recently has been more open to the ECHR and ECtHR judgments, allowing them to become indirectly binding.102

Concerns of an anachronistic inward-turning trend of the EU legal order were confirmed by Opinion 2/13. Here the CJEU revealed a parochial turn in its conception of EU human rights which seems to be dissociated from their international counterparts. To an extent it seems to miss the point of an external human rights accountability and control in limiting public power and authority exercised by the EU almost entirely. Opinion 2/13 marks a provisional climax of this trend in rejecting any formalisation of the relationship with the ECHR, in spite of the mandate by the ‘masters of the Treaties’, reflecting a renewed reluctance of the CJEU to subject EU law and itself to the jurisdiction of the European Court of Human Rights – a trend echoed in regard to other tribunals.103 The Court, in part under the guise of ‘protecting’ EU fundamental rights from external influences, also risks putting fundamental rights under a general reservation of ‘autonomy’ of the EU and the caveat of ‘realising’ them ‘within the framework of the structure and objectives of the EU.’104 Read in a pessimistic way, this could pave the way to justifying the ‘economic’ bias of the EU in relation to fundamental rights in the EU’s human rights architecture,105 and thus profoundly change normative hierarchies in the EU legal order.

The rejection of accession itself runs counter to an incremental constitutionalist rationale, but if not explained just by self-interest, it could be explained by a much more dramatic ‘constitutional moment’, in which the CJEU in fact asserts sovereignty (under the guise of the label of ‘autonomy’) against international law,106 but indirectly also against the Member States. Such a claim can

100 For an argument that the CJEU is delaying accession to the ECHR in order not to expose Charter jurisprudence in its formation to an ‘overwhelming’ influence of the ECHR see Adam Lazowski and Ramses A Wessel, ‘When Caveats Turn Into Locks: Opinion 2/13 on Accession of the European Union to the ECHR’ (2015) 16 German Law Journal 179, 190 et passim.101 Entscheidungen des Bundesverfassungsgerichts (BVerfGE) 111, 307 (2004) – Görgülü; see also Görgülü v Germany, App no 74969/01 (ECHR, 26 February 2004).102 BVerfGE 128, 326 (2011) – Preventive Detention; see also the analysis by Julia Rackow, ‘From Conflict to Cooperation: The Relationship Between Karlsruhe and Strasbourg’ in Katja S Ziegler, Elizabeth Wicks and Loveday Hodson (eds), The UK and European Human Rights: A Strained Relationship? (Hart Publishing 2015), 380, 388.103 The Commission (in amicus curiae submissions) has argued recently in relation to a number of ICSID cases concerning intra-EU BITS, that ICSID should decline jurisdiction over issue which inevitably would involve EU law: AES Summit Generation Limited and another v Republic of Hungary, Award, 23 September 2010, ICSID Case no ARB/07/22, para 8.2; Electrabel S.A (Belgium) v Republic of Hungary, Decision on Jurisdiction, Applicable Law and Liability, 30 November 2012, ICSID Case no ARB/07/19, para 5.20; Ioan Micula and others v Romania, Award, 11 December 2013, ICSID Case no ARB/05/20, paras 316-317.104 Opinion 2/13 (n 7), para 170.105 Katja S Ziegler, ‘Grundfreiheiten und soziale Dimensionen des Binnenmarktes - die Verfassung als Impuls?’ (2004 (Beiheft 3)) Europarecht 13.106 In this sense indeed Jan Willem van Rossem, ‘The Autonomy of EU Law: More is Less?’ in Ramses Wessel and Steven Blockmans (eds), Between Autonomy and Dependence (Asser Press and Springer

18

only be assessed and validated in retrospect when its consequences and reactions become clearer. It could endanger the human rights equilibrium reached with the Member States and, as such, supremacy of EU law over national law. This is a question only for future evaluation. For the time being it can be said that Opinion 2/13 is protecting the CJEU’s interpretive monopoly in a far-reaching way, as well as attempting to protect the Court from the EU Member States against actions which are both remote and unlawful under EU law, and from a small possibility of ‘forum shopping’ in Strasbourg. By focusing on the internal constitutionalism through the lens of autonomy, two points are missed: that this very constitutionalism requires both an external human rights control mechanism – a dimension of the internal constitutional order and constitutionalisation process which seems entirely lost on the CJEU – and an appropriate human rights protection. The latter is still characterised by some significant gaps in the EU. The complacency about human rights protection in and by the EU is a more fundamental concern than the reluctance displayed by the CJEU in regard to potentially competing judicial fora.

3.3. The ECHR in a post-accession scenario

The relationship between EU law and the ECHR in a post-accession scenario is determined by, firstly, international law and, secondly, the general approach of EU law to international treaties, in particular their status within the EU legal order.

From the perspective of international law, the EU would subject itself formally to the international obligations following from the ECHR. This would be a novelty for the EU in the context of human rights, for even in the framework of ILO Conventions participation of the EU occurs through coordination of the Member States’ participation. But accession would be in the interest of the Member States which could be held liable for human rights violations not just in the EU Treaties,107 but also of acts of the EU,108 in particular in the case of diverging human rights standards between the EU and the ECHR. It would also at the same time be appropriate, given the advanced constitutionalism, and a further step of EU constitutionalism itself.

In regard to the reception of the ECHR in the internal legal order or the EU, it would normally be expected to follow the general rule, i.e. the ECHR would be expected to become an integral part and ‘as a rule, share the direct effect of EU law’.109 The ECHR contains subjective rights which would even pass the stricter criteria of the CJEU for direct effect in Intertanko.110 It would, within the scope of EU law and as EU law, also have primacy over all national law of the Member States.111 Even if one considers that direct effect cannot be taken for granted, the ECHR would still be able to inform the interpretation of the Charter.112

2013) 13, 25 f.107 See Matthews v United Kingdom App no 24833/94 (1999) 28 EHRR 361. See for more detailed discussion Tobias Lock, ‘Beyond Bosphorus: The European Court of Human Rights' Case Law on the Responsibility of Member States of International Organisations under the European Convention on Human Rights’ (2010) 10 European Human Rights Law Review 529 ff.108 See for example, Joined Cases 46/87 and 227/88 Hoechst AG v Commission [1989] ECR 2859. This is the position of principle, although it has been softened in several respects, cf Bosphorus v Ireland App no 45036/98 (GC) (2006) 42 EHRR 1; Behrami v France and Saramati v France et al no 71312/01 and 78166/01(2007) 45 EHRR SE10. 109 See View of AG Kokott, Opinion 2/13 (n 7), para 198.110 Intertanko (n 8), paras 59 ff.111 Opinion 2/13 (n 7), para 199.112 But see above text around n 99.

19

The hierarchy of the ECHR within EU law is more difficult to establish. As a treaty it would generally be ranked above secondary but below primary EU law. However, the ECHR also enjoys already before a formal accession a special constitutional significance and status113 in the EU legal order. It is part of the human rights acquis as general principles referred to in Article 6(3) TEU, which implies a formal rank at the level of primary EU law. Its special status is further highlighted, at least indirectly, by the accession criteria (Articles 49 and 2 TEU) and the mechanism in Article 7 TEU. However, there is by virtue of the TFEU a potentially higher rank above primary EU law of (some) general principles of EU law (of which the ECHR partakes). The CJEU established in Kadi I a hierarchy within primary law which can be derived from the reasoning that there are certain rules of primary law from which there can be no derogation. (And which therefore also limit the exceptions contained in the ‘ordinary’ treaty rules of Article 351 TFEU and Article 347 TFEU, on the basis of which, for example, obligations of the Member States under the UN Charter could have been given priority over EU primary law). It counted among these fundamental rules ‘the principles of liberty, democracy and respect for human rights and fundamental freedoms enshrined in Article [6(3) TEU] as a foundation of the Union.’114 Thus, substantively, via the prism of general principles of EU law, the ECHR forms part of the foundational principles of the EU treaties. This is even acknowledged, albeit in a softer form only through the lens of interpretation by the CJEU in Opinion 2/13: ‘ECHR must be taken into account in the interpretation and application of EU primary law […] and ‘a careful balance must always be struck between those fundamental rights and the relevant provisions of primary law.’115

Finally, it has been debated whether the application of the so-called Bosphorus presumption116 will continue to be justified, according to which EU law is presumed to comply with the ECHR and as a result of which the ECtHR restricts its review. In a post-accession scenario, such a presumption would conflict with the purpose of an external human rights supervision in a hierarchical (rather than currently pluralistic) setting as well as with the equality of the parties to the ECHR in relation to the obligations undertaken. There should be no continued room for a presumption of compliance.

To conclude, it may be asked what impact EU accession to the ECHR will have. Irrespective of the internal status and application of the ECHR in the wider sense in the EU, accession would streamline the human rights architecture in the EU, avoiding double-bind situations for the Member States, fill existing gaps and, to an extent, it may be expected to harmonise standards. Interaction of legal orders is not entirely without risks. It cuts both ways and may lead to a levelling up or a levelling down, but an openness to interaction inherently also bears the chance to influence and shape standards.

4. The autonomy focus of the CJEU

113 Opinion 2/13 (n 7), para 201 ff.114 Kadi I (n 8), para 303, see also paras 282 ff, 304 ff. Katja S Ziegler, ‘Strengthening the Rule of Law, but Fragmenting International Law: The Kadi Decision of the ECJ from the Perspective of Human Rights’ (2009) 9 Human Rights Law Review 288, 297.115 Opinion 2/13 (n 7), para 204.116 Bosphorus v Ireland (n 108), paras 152 ff, esp 154-156; cf also Robert Uerpmann-Wittzack, ‘Die Grundrechtskontrolle durch den EGMR nach dem Beitritt der EU’ (2013) 68 Zeitschrift für Öffentliches Recht 519, 524 with further references.

20

One underlying reason for the Court to tighten the gatekeeping criteria towards international law is an increasing concern for, and emphasis of, the ‘autonomy’ of the EU legal order. The preliminary high water mark of this autonomy focus has been the Court’s Opinion 2/13 of 18 December 2014 which expresses this prominently: ‘The autonomy enjoyed by EU law in relation to the laws of the Member States and in relation to international law requires that the interpretation of those fundamental rights be ensured within the framework of the structure and objectives of the EU.’117

As will be shown, this approach reflects a conception of autonomy which puts the relationship between EU law and international law, including international human rights on its head. Autonomy of the EU legal order in a very far-reaching way becomes the all-encompassing reason why a certain protection of human rights is constructed rather than grounding rights protection in the intrinsic value(s) of rights. This is hardly short of putting the relationship of EU legal order with the international one under a general reservation of EU law. As mentioned above, it could also have implications for the role, status and rank of human rights within the EU legal order.118

This focus on autonomy is at the same time surprising and unsurprising: surprising, because the EU is not a state, as the CJEU emphasised itself repeatedly,119 and autonomy is traditionally a concern of states – usually as a manifestation of sovereignty; surprising also, as it runs counter to the trend regarding states whose sovereignty has been eroded or ‘overcome’ in many respects. The EU itself is driver and expression of this trend.

At the same time, such an approach is also unsurprising, as it follows one reading of the logic of EU constitutionalism if it is considered to ‘track’ the evolution towards a state-like entity and building an internal, constitutional protection of human rights.

Hence there is something ironic, in addition to being anachronistic, about the EU increasingly adopting this very stance, while it depends on the Member States relaxing similar approaches, and it is not a risk-free strategy of the EU. There is also a tension inherent in linking the EU’s non-state character with the conclusion that therefore there should be less external accountability. The Court in rejecting an external human rights control may consider it unnecessary, as perhaps it perceives the EU as an international organisation as less prone to degenerate into nationalism and other aberrations to which (some) states have shown themselves susceptible in the past.120 If anything, starting from the idea that any holding of power and authority can be abused, there should be more external accountability, precisely because the EU is not a state where there are still more legitimation and accountability mechanisms at play which complement one another. Or the approach reveals a conflation of internal and external conceptions of human rights. EU human rights are, of course, for the individual invoking them, international human rights at the same level as the ECHR. The ECtHR in certain defined scenarios is in competition with the CJEU when protecting individual rights. But EU human rights are not comprehensive human rights, most notably in regard to the Member States, for which such a conceptualisation, if adopted, should ring alarm bells of ‘competence creep’. From the perspective of the EU as the object of control, EU human rights are an internal control mechanism, which means there is still a need for external supervision. 117 Opinion 2/13 (n 7), para 170.118 See above text around n 104.119 Opinion 2/13 (n 7), para 49, 159, 193.120 Turkuler Isiksel, ‘European Exceptionalism and the EU's Accession to the ECHR’, Max Weber Conference on 'The Power of Constitutional Courts in a Global World', New York University, 9-10 April 2015, available at SSRN: <http://ssrncom/abstract=2590178> 10 accessed 22 July 2015.

21

The following section explores what it means to speak of the ‘autonomy’ of the EU legal order. It will be shown that the CJEU increasingly has used what is in essence a non-technical descriptive shorthand for various specific features of the EU in a normative way, turning it into an abstract ‘principle’ of a more general nature. ‘Autonomy’ as an abstract principle is then used as a flexible tool to ‘close’ the EU legal order.

4.1. What does ‘autonomy’ mean?

Opinion 2/13 can be criticised from a policy perspective for using the concept of autonomy inappropriately for various reasons: for example because it is untimely and against the trend that can be observed with states of a dissolution of sharp formal boundaries between legal orders and formal hierarchies; because it is inappropriate for a legal order that is part and dependent on international law; and because it is even counterproductive to the EU’s own trajectory of constitutionalism: it runs counter to the EU’s constitutional values and being subject to external human rights control/accountability might in itself be considered a feature of advanced constitutionalism; what the CJEU in effect demands is an exemption from international obligations that even sovereign and more directly democratically legitimated states take upon them; and it might even destabilise the constitutional equilibrium with the Member States (and the ECtHR) in the area of human rights under the Solange II/Bosphorus settlement,121 in particular as there are acknowledged deficiencies and gaps in the human rights protection of the EU. These are all arguments that can and have been made in response to Opinion 2/13.122

Beyond the question of the political appropriateness and timeliness of invoking autonomy lies the question whether the notion is valid as a matter of EU law as an independent legal concept, at least in the external dimension of the EU. Autonomy is not a term or concept of the Treaties. It is not defined by the Court in any detail. Rather, it is a claim derived from a free-standing teleological interpretation – without identifying a legitimate telos or its necessity. Even the partly underlying doctrines of direct effect and supremacy were more teleologically underpinned by safeguarding effectiveness of the EU legal order, something that cannot be equally said to motivate the creation of a principle of autonomy with regard to the international legal order. This would not matter too much as one might say that what rests on such shaky ground may easily be deconstructed. However, references to autonomy have developed their own dynamics and are on the trajectory to gain their independent normative force, as is well highlighted by Opinion 2/13. The following will take a step back and consider the different meanings and contexts in which the Court has used the notion of autonomy.

4.1.1. The internal dimension of autonomy: ‘A New Legal Order’

Autonomy is used as shorthand for what it is meant to safeguard in relation to the Member States: ‘the specific characteristics of the Union and Union law’. Such special characteristics, at least as a point of departure, would be a synonym for what is 121 German Federal Constitutional Court, Wünsche Handelsgesellschaft, 22 October 1986, BVerGE 73, 339, translation at [1987] 3 CMLR 225 (‘Solange II’); Bosphorus v Ireland (n 108), paras 152 ff, esp 154-156.122 See for example, debate on Opinion 2/13 (n 7) on <www.verfassungsblog.de > accessed 22 July 2015 and (2015) 16 German Law Journal, issue 1 at <https://www.germanlawjournal.com/index.php?pageID=2&vol=16&no=1> accessed 22 July 2015.

22

traditionally called a ‘new legal order of international law’ or simply ‘new legal order.’123 This refers to an internal dimension in relation to the Member States,124 characterised by direct effect and supremacy of EU law in the legal order of the Member States, individual rights and supranational law-making grounded on a division of competences between the EU and the Member States.125 In other words, these special features may be described as lex specialis features of EU law in relation to international law which may be what the ICJ referred to as a ‘certain autonomy’ of international organisations.126 This is indeed also the point of departure of the CJEU in Opinion 2/13.127

‘As the Court of Justice has repeatedly held, the founding treaties of the EU, unlike ordinary international treaties, established a new legal order, possessing its own institutions, for the benefit of which the Member States thereof have limited their sovereign rights, in ever wider fields, and the subjects of which comprise not only those States but also their nationals (see, in particular, judgments in van Gend & Loos, 26/62, EU:C:1963:1, p. 12, and Costa, 6/64, EU:C:1964:66, p. 593, and Opinion 1/09, EU:C:2011:123, paragraph 65). 158. The fact that the EU has a new kind of legal order, the nature of which is peculiar to the EU, its own constitutional framework and founding principles, a particularly sophisticated institutional structure and a full set of legal rules to ensure its operation, has consequences as regards the procedure for and conditions of accession to the ECHR.’128

From the classical case law as well as Opinion 2/13 it may be extracted that the specific feature is the quasi-federal but not federal-hierarchical but federal-pluralist constitutional structure of the EU and its Member States which is frequently described as being sui generis nature of the EU: the EU and its Member States can be equated neither with a state with single legal personality under international law, nor with an international organisation. This is the essence of what presents itself in a number of technical problems of ECHR accession: a fully federal state would be representing and be responsible for breaches of international law by its components. The situation is more complicated with the EU as the Member States remain subjects of international law and also parties to the ECHR, thus ‘internationalising’ what would be constitutional problems of a federal state. The Draft Accession Agreement

123 The Court has dropped the reference to international law in its phrase the ‘new legal order of international law’ contained in van Gend (n 24) only a year later in Costa v ENEL (n 24). For an argument why this did not change anything substantively see de Witte, ‘European Union Law: How Autonomous is its Legal Order’ (n 21), 147.124 It has been pointed out that the CJEU did not consider the EU not to be an international organisation or ‘outside the scope of international law’, de Witte (n 21), 147.125 van Gend (n 24); Costa v ENEL ibid; Case 11/70 Internationale Handelsgesellschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel [1970] ECR 1125, para 3. René Barents, The Autonomy of Community Law (Kluwer Law International 2004); Theodor Schilling, ‘The Autonomy of the Community Legal Order: An Analysis of Possible Foundations’ (1996) 37 Harvard International Law Journal 389; JHH Weiler and Ulrich Haltern, ‘The Autonomy of the Community Legal Order - Through the Looking Glass’ (1996) 37 Harvard International Law Journal 411.126 Legality of the Use by A State of Nuclear Weapons in Armed Conflict, Advisory Opinion ICJ Rep 1996, para 19.127 Opinion 2/13 (n 7), paras 165-170, adding ‘a set of common values (para 168), see also point 3 below.128 ibid, paras 157 f.

23

addresses the special nature of the EU as a non-state entity and in relation to the Member States in various ways, most notably by the co-respondent mechanism.

4.1.2. The external dimension of autonomy

In the external dimension, Opinion 1/00 on the establishment of a European Common Aviation Area listed two aspects of autonomy of the Community legal order:

‘Preservation of the autonomy of the Community legal order requires therefore, first, that the essential character of the powers of the Community and its institutions as conceived in the Treaty remain unaltered (Opinion 1/91, paragraphs 61 to 65, and 1/92, paragraphs 32 and 41).13. Second, it requires that the procedures for ensuring uniform interpretation of the rules of the ECAA Agreement and for resolving disputes will not have the effect of binding the Community and its institutions, in the exercise of their internal powers, to a particular interpretation of the rules of Community law referred to in that agreement (Opinions 1/91 and 1/92).’129

‘Essential character of powers of the Community’ may equated to the ‘specific characteristics’ of EU law in relation to the Member States, as described in point 1, but limited by a requirement of essentiality, however that may be interpreted.130 Additionally, in the external dimension the position in regard to other international dispute settlement bodies becomes relevant: an international agreement may not give an external dispute settlement body the power to interpret EU law authoritatively in regard to the internal application in the EU. From these two requirements it may be concluded that not all EU rules are non-negotiable in the context of an international agreement (only ‘essential’ ones), and that another dispute settlement body is not barred from considering EU law, it just must not intend to make internally binding pronouncements on their interpretation.

4.1.3. Autonomy and fundamental values

Autonomy may also been linked further to what it is meant to safeguard, but with a narrower protective thrust than any of the ‘specific characteristics’ discussed in point 1 above: ‘autonomy’ protects certain overarching values of the EU legal order such as the rule of law and ‘fundamental rights which are an integral part of the general principles of European Union law.’131 The reference is reminiscent of what amounts to the untouchable core of the German Constitution.132 This was most prominent in Kadi II. This variant can still be subsumed under the ‘essential character’ (point 2) or ‘specific characteristics’ (point 1).133 It has to be emphasised, however, that in contrast to Kadi I,134 Kadi II does not mention or even cross-reference ‘autonomy’ of the EU or the internal legal order of the EU, but was able to put into the foreground what was

129 Opinion 1/00 [2002] ECR I-3498, paras 12 f (emphasis added).130 See for one possibility section 3. Immediately below.131 Kadi II (n 38), paras 66 f.132 Cf Article 79(3) of the German Constitution and BVerfGE 123, 267, 358 (2009) – Lisbon Treaty.133 In this sense also van Rossem (n 106), 18.134 See eg Kadi I (n 8), para 316: ‘the review by the Court of the validity of any Community measure in the light of fundamental rights must be considered to be the expression, in a community based on the rule of law, of a constitutional guarantee stemming from the EC Treaty as an autonomous legal system which is not to be prejudiced by an international agreement.’

24

only a supporting obiter (but more convincing than the actual basis of the judgment) in Kadi I135, i.e. non-derogable constitutional principles of the EU: ‘the principles of liberty, democracy and respect for human rights’.136 The Court in Kadi II (rightly) relied on considerations of hierarchy of specific constitutional norms/fundamental rights, rather than a more general and abstract reasoning based on autonomy/dualism displayed in Kadi I (see Section 4.1.4. immediately below), while still being able to stay within the confines of the de facto precedent of Kadi I.137

4.1.4. Autonomy as a monopoly of jurisdiction of the CJEU

Autonomy is also referred to as a principle to protecting the CJEU’s jurisdiction in relation to other international tribunals (MOX, Opinion 2/94, Opinion 1/09, Opinion 2/13).138 It must be pointed out that traditionally, as reflected in the MOX case, the threshold for when the autonomy of jurisdiction was threatened was much higher than in Opinion 2/13. The CJEU required in MOX only that another agreement must not preclude compliance of EU Member States with what is now Art 344 TFEU, i.e. the mere possibility that Member States might bring an action was not relevant for assessing the compatibility of an international agreement with EU law.139 Opinion 2/13 required more, namely that an international agreement does not even create the possibility of Member States taking recourse to a different tribunal in the first place. This shifts the burden of compliance with EU law from the Member States to the level of the international treaty, displaying serious distrust in the Member States and lack of confidence in the EU’s own constitutional rules and enforcement mechanism!140

In addition, Opinion 2/13 also requires more for international dispute settlement mechanisms to be compatible with the treaty. In effect it bars other courts and tribunals from the mere consideration of EU law even as a fact, for example when assessing whether the EU, as any state, is in breach of the ECHR. Although this is not stated by the Court as a principle, it is the practical upshot of its very strict approach to the DAA.141

4.1.5. Autonomy as a synonym for dualism towards international law

135 Ibid (n 8), on the basis of paras 281-285, 303 of the judgment. Kadi I is not unequivocal in its reliance on dualism, see for example, de Witte (n 21), 154.136 Kadi I (n 8), para 303, see also paras 282 ff, 304 ff. 137 Kadi II (n 38), para 66. The Court which was invited by Council, Commission and several interveners (for example, UK) to revisit the dualist approach adopted in Kadi I decided with surprising brevity of reasoning, merely referring explicitly to the normative hierarchy argument relating to fundamental constitutional values, and not the autonomy argument.138 C-459/03 Commission v Ireland (MOX plant) [2006] ECR I-4657; Opinion 2/94 (n 86); Opinion 1/09 [2011] ECR I-1137; Opinion 2/13 (n 7).139 MOX (n 138), paras 122 ff, 124.140 In contrast to the View of AG Kokott in Opinion 2/13 (n 7), paras 114 ff who considered the infringement procedure or interim measures to be sufficient safeguards. For an example of a successful implementation of the duty of loyal cooperation in Art 4 (3) TEU (previously Art 10 TEC) via and infringement action see, for example, C-45/07 Commission v Hellenic Republic [2009] ECR I-701. See also for an extensive interpretation of Union powers in external relations of the Member States C-399/12 Germany v Council, ECLI:EU:C:2014:2258 which has raised concern about the EU affecting the efficiency of international organisations. 141 See for details about this criticism Piet Eeckhout, ‘Opinion 2/13 on EU Accession to the ECHR and Judicial Dialogue - Autonomy or Autarky?’ [2015] Jean Monnet Working Paper 01/15; Marten Breuer, ‘"Wasch mir den Pelz, aber mach mich nicht nass!" Das zweite Gutachten des EuGHzum EMRK-Beitritt der Europäischen Union’ (2015) Europarecht 330 ff, 346.

25

Autonomy of the EU legal order has been seen more generally as a synonym for a dualist approach to international law. A less benign reading of the reference in Kadi I to an ‘internal and autonomous legal order’142 sees the ratio decidendi of the judgment in the separateness of the legal orders – rather than the values at stake.143 On the basis of separateness the CJEU was able to decide the case purely on the basis of reviewing a domestic act by the (constitutional) human rights standards of EU law, excluding international obligations as irrelevant from the case,144 and ruling out any direct review of the Security Council Resolutions, even for breaches of ius cogens at the same time.145 This latter aspect is difficult to reconcile with the value-based approach (Section 4.1.3. above) and weakens the emphasis on the own constitutional values in Kadi I itself. The result of Kadi I can hardly be criticised, particularly as accountability is both very thin at international level and significantly taken back at national level in relation to executive power in foreign affairs.146 But even though the Court creates such accountability in Kadi I and even more so in Kadi II,147 a dualist reasoning, focused on autonomy is unnecessarily antagonistic to general international law and the international legal order.148 Autonomy as dualism, in contrast to all meanings discussed so far, does not consider the substance but is a purely formalistic shield.

4.1.6. Autonomy as an overarching legal principle: from myth to reality?

Thus far, apart from the variant in which autonomy is a potential synonym for dualism, autonomy was a limited concept referring to either specific substantive characteristics of the EU legal order (the constitutional arrangement with the Member States and substantive values) or the autonomy over issuing binding interpretation of EU law internally.

Opinion 2/13, however, constructs autonomy akin to an independent or overarching legal principle or even general principle of EU law which becomes shorthand for the entire legal order of the EU in Opinion 2/13,149 distinguished from the external dimension of state sovereignty only by name, in spite of the Court’s emphasis that it is not a state.150

With regard to the ECHR DAA, the CJEU seems to go beyond both criteria mentioned in point 2 (‘essential character’/’specific characteristics’) and in effect, making all of EU law (and not just ‘essential’ characteristics) sacrosanct. ‘Mutual trust between the Member States’ is elevated to a principle ‘of fundamental

142 Kadi I (n 8), para 317 f. 143 Ibid. See for analysis of this aspect on the one hand de Witte (n 21), 147, 154; on the other hand Ziegler (n 114), 293.144 Peter Hilpold, ‘EU Law and UN Law in Conflict: The Kadi Case’ (2009) 13 Max Planck Yearbook of United Nations Law 141, 160; Ziegler (n 7), 289 ff; Ziegler (n 40).145 Kadi I (n 8), para 287.146 The Kafkaesque factual scenarios of some of the anti-terrorism sanction cases demostrate how the UN Al-Quaida sanctions mechanism may led to a total lack of accountability of national executives involved.147 Kadi I (n 8); Kadi II (n 38).148 See in more detail Ziegler (notes 7, 40, 114). For an ex-post argument that Kadi I did not lead to a lack of compliance in the particular case of the UN sanctions regime see Joris Larik, ‘The Kadi Saga as a Tale of ’Strict Observance’ of International Law: Obligations Under the UN Charter, Targeted Sanctions and Judicial Review in the European Union’ (2014) 61 Netherlands International Law Review 23, 26 ff.149 See quotation above in text at n 117.150 See above n 119.

26

importance’ along the way.151 This is already reflected in the more protective turn of the Court in regard to its jurisdiction by requiring the internationalisation of obligations of loyal cooperation and stricter boundaries in which other courts may consider EU law. The following passages in Opinion 2/13 further reveal the extensive meaning of the notion of autonomy, describing all substantive areas of EU law as ‘specific characteristics’. They could be read to suggest that the rationale of economic integration in essence precludes an external human rights supervision!

‘The autonomy enjoyed by EU law in relation to the laws of the Member States and in relation to international law requires that the interpretation of those fundamental rights be ensured within the framework of the structure and objectives of the EU (see, to that effect, judgments in Internationale Handelsgesellschaft, EU:C:1970:114, paragraph 4, and Kadi and Al Barakaat International Foundation v Council and Commission, EU:C:2008:461, paragraphs 281 to 285)… The pursuit of the EU’s objectives, as set out in Article 3 TEU, is entrusted to a series of fundamental provisions, such as those providing for the free movement of goods, services, capital and persons, citizenship of the Union, the area of freedom, security and justice, and competition policy. Those provisions, which are part of the framework of a system that is specific to the EU, are structured in such a way as to contribute  − each within its specific field and with its own particular characteristics − to the implementation of the process of integration that is the raison d’être of the EU itself… In order to ensure that the specific characteristics and the autonomy of that legal order are preserved, the Treaties have established a judicial system intended to ensure consistency and uniformity in the interpretation of EU law.’152

It is to be noted that the Court refers to ‘autonomy’ separately from the ‘specific characteristics’ in the above and more directly in regard to ECHR accession in the following statement: ‘that the accession of the EU to the ECHR as envisaged by the draft agreement is liable adversely to affect the specific characteristics of EU law and its autonomy.’153

In this formulation, which is repeated throughout Opinion 2/13, it is clear that autonomy is considered to be something additional to the specific characteristics. The Court does not refer to the specific characteristics which are the substance of autonomy. Rather, autonomy seems to be treated as a more abstract and overarching concept from which the additional consequence of incompatibility of the DAA is derived. There is a legitimate concern to protect the lex specialis character of EU law, in particular in relation to the Member States and more widely with respect to certain core values and the exclusive competence to interpret EU law in a binding way in the internal dimension. But there is no need to refer to these wider aspects as ‘autonomy’ of the EU legal order. In fact, the CJEU has not always invoked ‘autonomy’ in such contexts.154

The shift to an abstract ‘principle’ of autonomy is a dangerous precedent likely to be used to justify whatever restrictive approach to international law is considered appropriate. With the same argument (but label of sovereignty) states have used the

151 Opinion 2/13, para 191.152 Ibid (n 7), paras 170, 172,174 (emphasis added).153 Opinion 2/13 (n 7), para 200 (emphasis added).154 Kadi II (n 38), para 66; Diakité (n 8); de Witte (n 21), 153.

27

‘autonomy’ of their domestic legal orders to try to shield them from international obligations. This goes as far as to, in effect, describe the lacunae of judicial review and hence human rights protection in the CFSP as a ‘specific characteristic’ of EU law which is to be protected!155 It may also be asked what the consequences for the internal protection of fundamental rights may be if they are seemingly subordinated to a ‘the framework of the structure and objectives of the EU.’156 It may also be a principle that may be turned against the Member States in a more far-reaching way than is the case at present.

4.2. Too extensive interpretation of the accession conditions

The wide and abstract conception of autonomy influences the Court’s interpretation of the accession conditions, leading to an overly restrictive approach.

EU accession to the ECHR could be considered simply as an obligation of the EU created by the ‘Masters of the Treaties’ against the backdrop of Opinion 2/94. If accession is considered to be conditional on meeting certain criteria (reflected in Article 6(2) TEU, the Declaration on Article 6(2) and Protocol 8 to the Treaty of Lisbon), these criteria are subject to interpretation, also in the light of whatever significance and meaning is attributed to the notion of autonomy. It will be argued in the following that the Court interpreted these accession conditions157 set by the Member States across the board too ‘protectively’, making accession to the ECHR at the very least extremely difficult. Such an interpretation raises the question whether the conditions as interpreted by the CJEU are compatible with the object and purpose of the ECHR as a system to protect human rights to which the Member States wanted to subject the EU as well as the intent of the parties. It will be argued that the accession conditions must be interpreted narrowly in the light of the Member States’ concerns about ECHR accession which the conditions sought to address. These concerns were mainly to safeguard Member States’ competences. However, throughout the accession process a certain change in attitude and in the dynamics of the process is noticeable, shifting away from Member States’ concerns to the concerns of the EU and CJEU.158 Even though the EU’s and CJEU’s concerns found extensive accommodation in the DAA, they culminated in Opinion 2/13.

Looking at the reasons of the Court in turn why it considers specific characteristics of the EU to be violated by the DAA, it is argued that the Court has adopted an overshooting restrictive interpretation of the accession conditions, based on autonomy:

Firstly, the CJEU is concerned that a conflict between Article 53 ECHR and Article 53 of the Charter would ‘compromise the level of protection provided for by the Charter or the primacy, unity and effectiveness of EU law’159. Article 53 ECHR 155 Opinion 2/13 (n 7), para 249 ff, cf heading before para 249.156 See already above text around n 105.157 Cf Opinion 2/13, paras 159 ff.158 See Sonia Morano-Foadi and Stelios Andreadakis, A Report on the Protection of Fundamental Rights in Europe: A Reflection on the Relationship between the Court of Justice of the European Union and the European Court of Human Rights post Lisbon <https://dmcoeint/CED20140017597> accessed 22 July 2015, 68 ff. The authors of the report interviewed CJEU judges in an early phase of the accession negotiations, revealing a generally welcoming attitude to accession; concerns expressed were addressed in the final DAA. See also Stelios Andreadakis, ‘Problems and Challenges of the EU’s Accession to the ECHR: Empirical Findings with a View to the Future’ in Sonia Morano-Foadi and Lucy Vickers (eds), Fundamental Rights in the EU A Matter for Two Courts (Hart Publishing 2015) 47 ff.159 Opinion 2/13, para 188; C-399/11 Stefano Melloni v Ministerio Fiscal, ECLI:EU:C:2013:107. .

28

allows for higher standards of protection by the Member States; Article 53 of the Charter prohibits, at least as interpreted by the CJEU in Melloni, that national human rights are more protective than EU human rights.

Eeckhout has pointed out that the CJEU adopts an unreasonably wide notion of conflict between EU treaty law and the ECHR and that such a conflict seems rather far-fetched.160 As long as EU law complies with the standards of the ECHR, it is possible for Member States to comply with both obligations. The ECHR allows, but does not require, a higher standard, allowing Member States to stay within the Melloni ruling. The argument of the CJEU seems so strained that it leads one to wonder whether the real concern is that the ECtHR will find that the EU falls short of the ECHR minimum and therefore requires the Member States (and the EU in a post-accession scenario) to raise standards. A ‘coordination’ of the ECHR to the Charter as the Court suggests161 (besides showing utter distrust towards the Member States) would be contrary to the purpose of external human rights supervision. Additionally, it may be pointed out that primacy, unity and effectiveness are, although fundamental concepts of EU law, not absolute. Hence it is unclear how these principles (as ‘specific characteristics’) are affected at all by any limitations that there may be as a result of accession to the ECHR. The standard should not be whether there is any adverse impact, however minor, but a clear violation of ‘specific characteristics’ – in the sense of constitutional fundamentals.

Secondly, the CJEU elevates ‘mutual trust between the Member States’ to a principle of ‘fundamental importance’ and then to a specific characteristic protected by autonomy which outbalances human rights protection.162 Mutual trust may well be the sociological underpinning of much – and a legal requirement of some – of EU law; but elevating it en passant to a fundamental constitutional principle, using it to shield the EU from human rights, in particular where the EU human rights protection has been found wanting,163 turns things on their head. At best this reflects a worrying misunderstanding not just of the function of external human rights, but also of internal (EU) human rights; at worst, it reflects a parochial isolationism or hubristic belief in its own infallibility.164

Thirdly, the CJEU tries in several regards to externalise its own internal relationship with the Member States, displaying a surprising distrust both in the Member States and in the EU enforcement mechanisms. To an extent such self-interest is understandable,165 but moulding jealousy (or institutional interest) into a legal argument, as the CJEU has done, leads to an unreasonable and unacceptable wide interpretation of accession conditions. Distrust of the Member States is at the core of several concerns of the CJEU: the concern about Member States adopting a higher level of rights protection in the Melloni-type situation (the Article 53 ECHR/the Charter issue discussed above166); the distrust in the Member States that they will go ‘forum shopping’ and use Protocol 16 to the ECHR inappropriately

160 Eeckhout, ‘Opinion 2/13 on EU Accession to the ECHR and Judicial Dialogue - Autonomy or Autarky?’ (n 141), p 13.161 Opinion 2/13 (n 7), para 189.162 ibid, para 191, 194.163 M.S.S. v Belgium and Greece ; Tarakhel v Switzerland App no 29217/12 (GC), ECHR 2014.164 Isiksel (n 120).165 See Daniel Halberstam, ‘"It's the Autonomy, Stupid!" A Modest Defense of Opinion 2/13 on EU Accession to the ECHR, and A Way Forward’ (2015) 16 German Law Journal 105; Christopher Krenn, ‘Autonomy and Effectiveness as Common Concerns: A Path to ECHR Accession After Opinion 2/13’ (2015) 16 German Law Journal 147166 See above text around n 140 ff.

29

(rather than the preliminary reference procedure under Article 267 TFEU), even though prior involvement mitigates that risk significantly; and finally the distrust of Member States to bring inter-state cases in the field of EU law in breach of Article 344 TFEU against other Member States in Strasbourg (even though the possibility exists without accession and hardly proved to be an issue167) which led the Court to interpret Article 344 TFEU as more limiting than was the case previously:168 ‘only the express exclusion of the ECtHR’s jurisdiction under Article 33 of the ECHR over disputes between Member States or between Member States and the EU in relation to the application of the ECHR within the scope ratione materiae of EU law would be compatible with Article 344 TFEU.’169

Fourthly, the Court adopts a ‘noli me tangere’ attitude, trying to prevent the ECtHR even looking at EU law when exercising its supervision in order to protect EU law from external influences and its own jurisdiction. The CJEU is concerned about the ECtHR interpreting EU law per se and implications this could have for the division of competences and allocation of responsibility between the Union and the Member States. In the context of the co-respondent mechanism the CJEU objects to the assessment by the ECtHR of ‘rules of EU law governing the division of powers between the EU and its Member States as well as the criteria for the attribution of their acts or omissions.’170 Two comments may be made: it is, on the one hand, an inevitable part of the ECtHR exercising its function to consider whether a rule of a legal order is compatible with the ECHR; in order to be able to do so, it must be able to interpret the rules. On the other hand, such consideration does not amount to an internally binding determination of competences or binding interpretation of EU law.171 This is quite a normal situation for any court applying a limited, defined standard of review, such as the ECtHR in regard to the ECHR or the German Federal Constitutional Court in regard to the Grundgesetz. Its task and jurisdiction is only to check whether an application of law violates the Grundgesetz, in particular its fundamental rights; it is not its task to function as a further instance of appeal and decide what would have been the correct application of the ordinary law in a comprehensive sense.172 To be sure, transgressions arguably may happen as the boundary is not always clearly cut. But this is insufficient reason to totally exclude any supervision; moreover, it is unclear why it should ‘jeopardize the autonomy’ of EU law.173 In regard to the rules of responsibility this could only be even argued if the CJEU considers responsibility for breach of an international obligation to be subject to special EU rules (which would be hard to sustain). The question of attribution of a breach of international law is based ultimately on factual determination of who acted and committed (caused) the violation and not which entity had the competence to act.174

167 As demonstrated by the MOX case (n 138).168 ibid. See above text following n 138.169 Opinion 2/13 (n 7), para 213.170 ibid, para 224.171 Lazowski and Wessel (n 100), 199; Eeckhout (n 141), p 28.172 Constant jurisprudence, see for example Bundesverfassungsgericht, 2 BvR 1821/99, ECLI:DE:BVerG:1999rk19991024.2bvr182199, para 5; BVerfGE 18, 85, 92 f (1964).173 Sionaidh Douglas-Scott, ‘The Relationship between the EU and the ECHR Five Years on From the Treaty of Lisbon’ in Ulf Bernitz, Sybe A de Vries and Stephen Weatherill (eds), Five Years Legally Binding Charter of Fundamental Rights (Hart Publishing forthcoming), Oxford Legal Studies Research Paper <http://ssrn.com/abstract=2533207 > , fn 12, accessed 22 July 2015.174 Giorgio Gaja, ‘The "Co-Respondent" Mechanisms' According to the Draft Agreement for the Accession of the EU to the ECHR’ in Vasiliki Kosta, Nikos Skoutaris and Vassilis Tzevelekos (eds), The EU Accession to the ECHR (Hart 2015) 341, 346.

30

Finally, as mentioned before, the CJEU elevates a structural deficiency of EU law in regard to judicial review in the area of CFSP to a specific feature of EU law to justify that also external human rights supervision should be restricted. This seems counterintuitive both from the perspective of the EU’s own constitutional values and from the perspective of the transformative potential of such external review to incentivise Member States to create jurisdiction of the CJEU. Moreover, supranational features (and hence the ‘autonomy’) of the EU would not be affected in this area.175 Also, it hardly prevents such review, given that the relevant acts are likely to be mostly Member States’ acts which are subject to review by Member States’ courts (and the ECtHR), whether the EU is party to the ECHR or not.176

4.3. A more restricted reading of the accession conditions in the light of the intent of the member states and purpose of the conditions

Inherent in these specific criticisms is the wider criticism that the Court distorted the conditions in Article 6(2) TEU, Protocol 8 and the Declaration on Article 6(2). Interpreting the accession conditions in the context of their drafting, the intent of the Member States as well as their purpose, a restricted (narrow) reading of the conditions must be adopted.

Article 6(2) TEU – to be seen in the light of Opinion 2/94 – creates the legal basis and expresses the Member States’ intention for the EU to accede. In paving the way to accession Article 6(2) TEU and in particular Protocol 8 of the Lisbon Treaty deal primarily with two issues: first, Member States’ issues and concerns about accession, especially the impact on the competences of the Union in relation to the Member States; and, secondly, ‘technicalities’ resulting from the accession of the EU as a non-state entity, and a fortiori, non-federal state party to the ECHR (and that the EU, unlike a federal state could not be attributed automatically responsibility for the actions of its component parts).

Article 1 of Protocol 8 requires ‘provision for preserving the specific characteristics of the Union and Union law’, making particular reference to procedural issues and participation in the control bodies. This concern of the Member States is reflected in Article 6(2) TEU itself which provides that ‘accession shall not affect the Union’s competences as defined in the Treaties’. Constitutionally, competences of the EU define the relationship between the Member States and the Union. Article 2 of Protocol 8 echoes this, restating and clarifying further that the Accession Agreement ‘shall ensure that nothing therein affects the situation of Member States in relation to the’ ECHR.

The Declaration on Article 6(2) to the Treaty of Lisbon177 takes a softer formulation in stating that accession should ‘preserve the specific features of Union law’ (noting an existing dialogue between the CJEU and the ECtHR which should be reinforced post-accession). 175 See View of AG Kokott in Opinion 2/13 (n 7), para 192n.176 See cases cited above n 108. Christophe Hillion, ‘A Powerless Court? The European Court of Justice and the Common Foreign and Security Policy’ in Marise Cremona and Anne Thies (eds), The European Court of Justice and External Relations Law (Hart Publishing 2014) 47, 67.177 Declaration no 2 to the Lisbon Treaty: ‘Declaration on Article 6(2) of the Treaty on European Union. The Conference agrees that the Union's accession to the European Convention for the Protection of Human Rights and Fundamental Freedoms should be arranged in such a way as to preserve the specific features of Union law. In this connection, the Conference notes the existence of a regular dialogue between the Court of Justice of the European Union and the European Court of Human Rights; such dialogue could be reinforced when the Union accedes to that Convention.’

31

In the light of the context of the framework in the EU treaties, including Protocol 8, what is meant here is not any competence of the Union. Such an interpretation would be nonsensical and defeat the main purpose of accession (as well as the object and purpose of the ECHR, cf Article 19 lit. c) of the Vienna Convention on the Law of Treaties). The point and purpose of subjecting the EU to an external (human rights) supervision is precisely to limit the power of the EU (as is the case with any party to the ECHR) which impacts inevitably to a degree on the exercise of its competence. What is reflected here, and elsewhere in the amendments made by the Lisbon Treaty in the context of human rights, i.e. the Charter of Fundamental Rights, for example, Article 51(2) of the Charter, is the concern of the Member States about a ‘competence creep’ of the EU in the area of human rights. This concern of the Member States is not totally unfounded for two reasons:

Firstly, the CJEU has a track record of a skewed application of EU human rights. It has been criticised for being more inclined to apply EU human rights over Member States’ acts in the scope (or even mere context of EU law) than annulling its own acts. And it has at least in the past adopted a wide approach to the link required to EU law,178 thus allowing in effect a review of national acts by EU human rights. This is not the place to evaluate this jurisprudence, but merely to say that a concern from the Member States’ perspective about an expansion of EU competence in the area of human rights is not entirely without foundation.179

Secondly, the a post-accession human rights architecture is likely to give rise, in certain factual constellations, to a complex interaction of different layers of rights in the EU. This may lead to a de facto momentum or pressure for harmonisation of the different standards and a potential blurring of the boundaries of EU and Member States’ competences. In regard to the content of rights, it is not unlikely that there will be a gravitational effect to align interpretations of rights across different codifications. This can be triggered by the mere fact of interaction and engagement of the legal orders. But also from a mere practical point of view where a national official has to apply different sets of human rights in the context of EU law and in the context of national law, this will be likely to lead to interpreting different rights in the same way (perhaps allowing a public authority or court even to leave undecided the question of which body of human rights is applicable, a question which would require potentially complicated investigations into the competences at issue). If interpretations of different human rights codifications will in fact be aligned, Luxembourg will most certainly carry much weight in this regard.180 More specifically, in a post-accession scenario there is a

178 C-60/00 Mary Carpenter v Secretary of State for the Home Department [2002] ECR I-6279; Case C-617/10 Åklagaren v Hans Åkerberg Fransson, ECLI:EU:C:2013:280. Although more (overly) careful in regard to the application of the Charter recently in Case C-333/13 Elisabeta Dano and Florin Dano v Jobcenter Leipzig, ECLI:EU:C:2014:2358, 85 ff, 90.179 See also above n 97 for a similar concern in regard to the development of general principles of EU law by the Court.180 Such a concern is also expressed in Declaration no 53 to the Lisbon Treaty by the Czech Republic on the Charter of Fundamental Rights of the European Union, seeking to protect the ECHR from being watered down by EU law: ‘4. The Czech Republic further stresses that nothing in the Charter may be interpreted as restricting or adversely affecting human rights and fundamental freedoms as recognised, in their respective field of application, by Union law and by international agreements to which the Union or all the Member States are party, including the European Convention for the Protection of Human Rights and Fundamental Freedoms, and by the Member States' Constitutions.’

32

potential difference in regard to the status of the ECHR in the legal order of Member States in the scope of EU law (where it would likely be directly effective181) and in relation to Member States’ domestic law. The internal status of the ECHR within domestic law of the Member States varies depending on the formal status of treaties in the respective Member State’s legal order and the approaches of their courts.182 In practice the difference is likely to be not very significant because Member States will adopt various interpretative mechanism which will mitigate any differences. Structurally, from the perspective of competence, however, there is a possible basis for a Member States’ concern about a spill-over or de facto harmonisation of direct effect of the ECHR in the scope of EU law through the backdoor for domestic law.

The concern of a ‘competence creep’ of the EU in the area of human rights – and ECHR accession in particular – has been expressed by various Member States. The Explanatory Memorandum of the UK Foreign and Commonwealth Office (FCO) presented as written evidence to the House of Lords Select Committee on the Constitution UK in its inquiry on the implications of the Treaty of Lisbon for the UK 2007-8 confirms a reading of Protocol 8 as intended to protect Member States’ (rather than EU) competences:

‘All EU Member States are themselves parties to the ECHR. The Government has sought and achieved a legally binding Protocol that confirms that EU accession to the ECHR will not affect the situation of Member States in relation to the Convention—including the Protocols in which they participate, national derogations and reservations to the ECHR; nor increase the EU's competences. This states that: “. . . accession of the Union shall not affect the competences of the Union or the powers of its institutions. It shall ensure that nothing therein affects the situation of Member States in relation to the European Convention, in particular in relation to the Protocols thereto, measures taken by Member States derogating from the European Convention in accordance with Article 15 thereof and reservations to the European Convention made by Member States in accordance with Article 57 thereof”.’183

Or in other words, the concern of the Member States is not to expand EU competences. To limit EU competences simply was not a concern of the Member States, it is inherent in ECHR accession. The prominence of the Member States’ concern about a competence creep in the area of human rights in the treaty and Protocol is not significantly weakened by the fact that the Protocol highlights the existing obligation of Member States under EU law (Article 344 TFEU) to refrain from inter-state proceedings in the area of EU law in Strasbourg which is merely declaratory of the status quo.

Thus, there is a theme running through the legal framework of accession about safeguarding Member States’ competences against the EU. Ironically, the dynamics changed at some point in the accession process and negotiations, and concerns of the EU about accession (in particular characterised by distrust of its own Member States) gained sufficient voice to receive further accommodation in the Draft Accession 181 See above text around n 109.182 See above for the approach of the German Federal Constitutional Court, text around n 101 f.183 House of Lords Constitution Committee, EU (Amendment) Bill and the Lisbon Treaty: Implications for the UK Constitution Report of the 2007-8 Session (printed 19 March 2008) (emphasis added).

33

Agreement (for example, creation of the prior involvement procedure184). Member States’ concerns were only more or less indirectly represented through the lens of the EU – to the point that the division of competences between the EU and the Member States found only very one-sided reflection in Opinion 2/13, namely only in relation to Article 344 TFEU.185 The issue in essence is the ‘special nature’ of the EU in two respects: it is not a state viewed in itself, and it is not a federal state when viewed from the perspective of its component parts.

This background and interpretation of the accession conditions is relevant for the approach to the DAA. An overly restrictive approach, as taken by the CJEU in Opinion 2/13, in essence would be incompatible with the object and purpose of the TEU, Protocol and Declaration, and the intent of the ‘Masters of the Treaties’, or in other words, a breach of the Treaty mandate to accede.

Of course there is ultimately no arbiter of the question above the CJEU,186 so the issue can be settled only in a pluralistic setting. National constitutional courts could, for example, step up resistance again if the hubris of the CJEU is considered to reflect a disregard for fundamental rights that falls below the practical accommodation under the Solange II/Bosphorus settlement.187 This is unlikely to occur specifically with regard to the question of ECHR accession, not least because the question of membership to the ECHR is a rather abstract one. However, a more confrontational approach could easily be imagined for human rights violations in the context of the Common Asylum System, the CFSP or ASFJ – both from the Member States apex courts and the ECtHR.

5. Concluding remarks: hubris on a tightrope or an alternative engaged approach?

This chapter has taken the reader through the complex relationship between EU law and international law, analysing an increasing trend of closure of the former to the latter in the jurisprudence of the CJEU.188 It has more specifically shown that international human rights law, at least since the entry into force of the Charter, is no exception to this trend; not even if it is, like the ECHR, of European provenance and deeply intertwined with EU fundamental rights. In inquiring into the reasons for this, the chapter critically examined the prominent focus on autonomy in Opinion 2/13. Relating the notion of autonomy to earlier case law reveals a much expanded, overarching concept, potentially on the verge of being considered to be a general principle by the CJEU. Such a wide notion of autonomy distorts the interpretation of the accession conditions in the Article 6(2) TEU, Protocol 8 to the Treaty of Lisbon and the Declaration on Article 6(2) TEU. The chapter argued that the accession conditions should be interpreted narrowly in the light of the object and purpose and intention of the Member States when creating the obligation to accede. Member States

184 See Joint Communication from Presidents Costa and Skouris, 24 January 2011 at <www.curia.europa.eu/jcms/jcms/P_72317/ > accessed 22 July 2015 ; Jean-Paul Jacqué, ‘The Accession of the EU to the ECHR’ (2011) 48 Common Market Law Review 995, 1017 ff.185 See also Lazowski and Wessel (n 100), 186.186 Joseph HH Weiler, ‘Epilogue: Judging the Judges - Apology and Critique’ in Maurice Adams and others (eds), Judging Europe's Judges The Legitimacy of the Case Law of the European Court of Justice (Hart 2013) 235, 253.187 Above n 121.188 This contrasts with the engagement of the EU as an actor on the international scene, cf Jenö Czuczai, ‘The Autonomy of the EU Legal Order and the Law-Making Activities of International Organisations. Some Examples Regarding the Council's Most Recent Practice’ (2012) 31 Yearbook of European Law 452 ff.

34

were mainly preoccupied with a possible spill-over of the EU into their areas of competence which should inform the interpretation of the accession conditions.

The result of a restrictive approach to international law is that the EU legal order isolates itself from international influences and dynamics which would be particularly valuable for shaping and maintaining a modern system of human rights protection. More generally, isolating itself from international law is not just potentially damaging to international law, but could in the long term undermine the foundations of the EU itself. It may also undermine the legitimacy of the EU in its own constitutional setting in relation to the wider world, for example in regard to the human rights conditionality in treaties with third states. More specifically in regard to the attitude taken to ECHR accession, the hubris displayed by the Court may undermine the pluralistic Solange II/Bosphorus settlement.

Constitutionalism and internationalism are, however, not mutually exclusive! The international legal order looks favourably on the advancement of international law by ‘partial’ or specialised legal orders in that it allows for more specific rules and derogations from the general rule to an overwhelming extent (ius cogens excepted). In allowing such specialisation which mainly occurs through treaties, it does, however, require a certain process of open engagement with other international law. This follows already from the general principle of interpretation of treaties in good faith in Article 31(1) of the Vienna Convention on the Law of Treaties and more specifically from the so-called principle of systemic integration in Article 31(3) lit (c) VCLT which requires a treaty regime to take ‘into account together with the context … any relevant rules of international law applicable to the parties.’189 Article 31 VCLT is generally considered to reflect customary international law. 190 Such ‘systemic integration’ is an obligation, but a weak one. In essence it is a principle of contextual interpretation that requires engagement but does not prescribe that other rules necessarily prevail. But it is a mechanism that serves the special decentralised nature of the international legal order and customary international to avoid conflicts and to preserve a minimum of unity of different areas of international law. The decentralised nature means that the legal order is fundamentally challenged in its binding quality and hence existence by conflict with and breach of international rules. The process of engaged interpretation preserves a minimum of unity of international law as a system (not unity at the level of individual rules). If there is a potential conflict, interpretation may eliminate it, but if interpretation does not eliminate the conflict between norms of two legal orders, engagement turns into a potentially positive impulse for the development of the international legal order: where the reasons are explicit, they can contribute as opinio iuris to the development of other areas of (customary) international law. From this it follows that international organizations, their courts, and tribunals have a procedural duty and heightened responsibility in comparison to states to at least engage with other areas of international law and to attempt to avoid conflict in the interpretation of their own constitutional treaties, so as not to undermine fundamentally the system on which they are founded.

189 See for example, Campbell McLachlan, ‘The Principle of Systemic Integration and Article 31(3)(c) of the Vienna Convention ’ (2005) 54 International and Comparative Law Quarterly 279.190 Oliver Dörr, ‘Article 31: General Rule of Interpretation’ in Oliver Dörr and Kirsten Schmalenbach (eds), Vienna Convention on the Law of Treaties (Springer 2012) 521, para 6.

35